Paper Digest: SIGIR 2026 Papers & Highlights
SIGIR (Annual International ACM SIGIR Conference on Research and Development in Information Retrieval) is one of the top information retrieval conferences in the world. To help the community quickly catch up on the work presented in this conference, Paper Digest Team processed all accepted papers, and generated one highlight sentence (typically the main topic) for each paper. Readers are encouraged to read these machine generated highlights to quickly get the main idea of each paper.
To search for papers presented at SIGIR-2026 on a specific topic, please make use of the search by venue (SIGIR-2026) service. To summarize the latest research published at SIGIR-2026 on a specific topic, you can utilize the review by venue (SIGIR-2026) service. If you are interested in browsing papers by author, we have a comprehensive list of ~ 2,800 authors (SIGIR-2026). Additionally, you may want to explore our “Best Paper” Digest (SIGIR), which lists the most influential SIGIR papers since 1981.
Since 2018, Paper Digest has built a foundation of data spanning decades of conferences, journals, and research topics. The platform features a daily digest service that sifts through tens of thousands of new papers, clinical trials, news articles, and community posts, filtering the noise to highlight what matters most to specific interests. Beyond daily updates, dozens of built-in research tools streamline the academic workflow, supporting efficient reading and writing, comprehensive literature reviews, and automated research report generation.
Paper Digest Team
New York City, New York, 10017
team@paperdigest.org
TABLE 1: Paper Digest: SIGIR 2026 Papers & Highlights
| Paper | Author(s) | |
|---|---|---|
| 1 | Tool-Star: Empowering Multi-Tool Collaborative Web Agent Via Reinforcement Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we introduce Tool-Star, an end-to-end agentic post-training framework that empowers LLM-based web agents to strategically interact with external multi-tool environments. |
Guanting Dong; Yifei Chen; Xiaoxi Li; Jiajie Jin; Hongjin Qian; Yutao Zhu; Hangyu Mao; Guorui Zhou; Zhicheng Dou; Ji-Rong Wen; |
| 2 | SmartSearch: Process Reward-Guided Query Refinement for Search Agents Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As a result, the generated queries often remain inaccurate, leading to unexpected retrieval results and ultimately limiting search agents’ overall effectiveness. To mitigate this issue, we introduce SmartSearch, a framework built upon two key mechanisms: (1) Process Rewards, which provide fine-grained supervision for the quality of each intermediate search query through Dual-Level Credit Assessment. |
Tongyu Wen; Guanting Dong; Zhicheng Dou; |
| 3 | Retrieval-Augmented Contrastive Learning for Knowledge Tracing Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Knowledge Tracing (KT) models aim to predict student performance from interaction histories in order to support personalised learning. |
Kamal Berahmand; Mehrnoush Mohammadi; Homa Babai; Hassan Khosravi; |
| 4 | Retrieval-Augmented Contrastive Learning for Dynamic Graph Anomaly Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose DGRA-CL (Dynamic Graph Retrieval-Augmented Contrastive Learning), an unsupervised framework that learns discriminative temporal node representations for anomaly detection without labeled data. |
Kamal Berahmand; Saman Forouzandeh; Mehrnoush Mohammadi; Mahdi Jalili; |
| 5 | Modular Representation Compression: Adapting LLM Representations for Efficient and Effective Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Thus, we propose Modular Representation Compression (MARC) to explicitly control the modularity of LLMs. |
Yunjia Xi; Menghui Zhu; Jianghao Lin; Bo Chen; Ruiming Tang; Yong Yu; Weinan Zhang; |
| 6 | HiRA: Decoupling Planning and Execution with Hierarchical Reasoning in Deep Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we introduce HiRA, a hierarchical framework that separates strategic planning from specialized execution. |
Jiajie Jin; Xiaoxi Li; Yuyao Zhang; Guanting Dong; Zhao Yang; Yutao Zhu; Zhicheng Dou; |
| 7 | Internalizing Explicit Reasoning Into Latent Space for Dense Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose LaSER, a novel self-distillation framework that internalizes explicit reasoning into the latent space of dense retrievers. |
Jiajie Jin; Yanzhao Zhang; Mingxin Li; Dingkun Long; Pengjun Xie; Yutao Zhu; Zhicheng Dou; |
| 8 | Can QPP Choose The Right Query Variant? Evaluating Query Variant Selection for RAG Pipelines Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Large Language Models (LLMs) have made query reformulation ubiquitous in modern retrieval and Retrieval-Augmented Generation (RAG) pipelines, enabling the generation of multiple semantically equivalent query variants. |
Negar Arabzadeh; Andrew Drozdov; Michael Bendersky; Matei Zaharia; |
| 9 | TimelineReasoner: Advancing Timeline Summarization with Large Reasoning Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To leverage the reasoning capabilities of LRMs, we propose TimelineReasoner, a novel framework that shifts TLS from static generation to an active, reasoning-driven process. |
Liancheng Zhang; Xiaoxi Li; Zhicheng Dou; |
| 10 | Lighting The Way for BRIGHT: Reproducible Baselines with Anserini, Pyserini, and RankLLM Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present a systematic study of BRIGHT, a reasoning-focused retrieval benchmark, along with strong, reproducible reference methods integrated into Anserini, Pyserini, and RankLLM. |
Sahel Sharifymoghaddam; Yijun Ge; Raghav Vasudeva; Jimmy Lin; |
| 11 | ReST: A Plug-and-Play Spatially-Constrained Representation Enhancement Framework for Local-Life Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we propose ReST, a plug-and-play framework for Spatially-Constrained Representation Enhancement. |
Hao Jiang; Long Zhang; Guoquan Wang; Sheng Yu; Yang Zeng; Wencong Zeng; Fei Pan; Peng Jiang; Guorui Zhou; |
| 12 | Towards Mixed-Modal Retrieval for Universal Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we address the challenge of Universal Retrieval-Augmented Generation (URAG), which involves retrieving and reasoning over mixed-modal information to improve vision-language generation. |
Chenghao Zhang; Guanting Dong; Xinyu Yang; Zhicheng Dou; |
| 13 | Learning from Natural Language Feedback for Personalized Question Answering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce VAC, a novel framework for personalized response generation that replaces scalar rewards with natural language feedback (NLF) that are generated conditioned on the user profiles and the question narratives. |
Alireza Salemi; Hamed Zamani; |
| 14 | SurGE: A Benchmark and Evaluation Framework for Scientific Survey Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing evaluation methods typically rely on either human evaluation or custom metrics designed to validate specific pipelines, which restricts scalability and hinders fair comparison. To address this, we introduce SurGE, a benchmark and evaluation framework tailored for scientific survey generation. |
Weihang Su; Anzhe Xie; Qingyao Ai; Jianming Long; Xuanyi Chen; Jiaxin Mao; Ziyi Ye; Yiqun Liu; |
| 15 | Enhancing Judgment Document Generation Via Agentic Legal Information Collection and Rubric-Guided Optimization Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing approaches, typically relying on standard Retrieval-Augmented Generation and Supervised Fine-Tuning, often suffer from insufficient evidence recall, hallucinated statutory references, and logically flawed legal reasoning. To bridge this gap, we propose Judge-R1, a unified framework designed to enhance LLM-based judgment document generation by jointly improving legal information collection and judgment document generation. |
Weihang Su; Xuanyi Chen; Yueyue Wu; Qingyao Ai; Yiqun Liu; |
| 16 | C3Flow: SIMD-Style Concurrent Claude Code Workflow for Scaling Deep Research Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce C3Flow (Concurrent Claude Code Workflow), a framework that transforms Claude Code from an interactive assistant into a SIMD-style (Single Instruction, Multiple Data) concurrent compute engine. |
Yunfan Gao; Xinyi Huang; Yijie Zhong; Zhidong Fan; Zhengke Gui; Lei Liang; Yun Xiong; Haofen Wang; |
| 17 | Query-Aware Context Selection for Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we conduct an empirical study of how irrelevant retrieved passages affect downstream generation, analyzing their impact across multiple standard generator models. |
Maya Iratni; Mohand Boughanem; Taoufiq Dkaki; |
| 18 | Think, But Don’t Tell: Implicit Reasoning for LLM-based Sequential Recommendation Via Multi-Teacher Distillation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, deploying explicit CoT reasoning in real-world systems faces prohibitive challenges: (i) the conflict between the large model scale required for high-fidelity reasoning and the resource constraints of online services, and (ii) the excessive latency introduced by auto-regressive rationale generation. To address these issues, we propose I Reasoning via Multi-Teacher Distillation (IRMD), a novel framework that ‘compiles’ the reasoning abilities of large teacher LLMs into a lightweight student Small Language Model (SLM). |
Weihai Lu; Xiaoxi Cui; Chenke Yin; |
| 19 | Semantic IDs for Recommender Systems at Snapchat: Use Cases, Technical Challenges, and Design Choices Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: At Snapchat, we apply SIDs as auxiliary features for ranking models, and also explore SIDs as additional retrieval sources in different ML applications. In this paper, we discuss practical technical challenges we encountered while applying SIDs, experiments we have conducted, and design choices we have iterated to mitigate these challenges. |
Clark Mingxuan Ju; Tong Zhao; Leonardo Neves; Liam Collins; Bhuvesh Kumar; Jiwen Ren; Lili Zhang; Wenfeng Zhuo; Wen Zhang; Xiao Bai; Jinchao Li; Karthik Iyer; Zihao Fan; Yilun Xu; Yiwen Chen; Peicheng Yu; Manish Malik; Neil Shah; |
| 20 | LLM-EDT: Large Language Models Enhanced Cross-domain Sequential Recommendation with Dual-phase Training Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Thus, to address the aforementioned challenges, we propose an LLMs Enhanced Cross-domain Sequential Recommendation with Dual-phase Training (LLM-EDT). |
Ziwei Liu; Qidong Liu; Wanyu Wang; Yejing Wang; Pengyue Jia; Tong Xu; Wei Huang; Chong Chen; Xiangyu Zhao; |
| 21 | Uncertainty Quantification for Retrieval-Augmented Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Accurate estimation of UQ for RAR requires accounting for all sources of uncertainty, including those arising from retrieval and generation. In this paper, we account for these sources and introduce Retrieval-Augmented Reasoning Consistency (R2C), a novel UQ method for RAR. |
Heydar Soudani; Hamed Zamani; Faegheh Hasibi; |
| 22 | Equity Vs. Equality: Optimizing Ranking Fairness for Tailored Provider Needs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we introduce an equity-oriented fairness framework that explicitly models each provider’s preferences over key outcomes such as exposure and sales, thus evaluating whether a ranking algorithm can fulfill these individualized goals while maintaining overall fairness across providers. |
Yiteng Tu; Weihang Su; Shuguang Han; Yiqun Liu; Qingyao Ai; |
| 23 | Analytical Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As a result, they struggle to support analytical queries that have diverse utility concepts and high accountability requirements. In this paper, we propose analytical search as a distinct and emerging search paradigm designed to fulfill these analytical information needs. |
Yiteng Tu; Shuo Miao; Weihang Su; Yiqun Liu; Qingyao Ai; |
| 24 | Generation-Augmented Video Corpus Moment Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing discriminative approaches typically rely on shallow visual-textual feature matching mechanisms, which often struggle to capture fine-grained semantic differences. To address this limitation, we propose Video-GAR, a novel framework that reframes the conventional retrieval task from superficial matching to generative understanding, positing that the capability for query reconstruction evidences deep semantic comprehension. |
Mingjin Kuai; Qianyin Xiao; Juncheng Li; Jin Peng; Lizi Liao; Wei Ji; |
| 25 | Jina-embeddings-v5-text: Compact and Robust Text Embedding Models Using Task-Targeted Distillation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present a pair of new embedding models trained with a novel training regimen that combines model distillation with task-specific contrastive losses. |
Mohammad Kalim Akram; Saba Sturua; Nastia Havriushenko; Quentin Herreros; Michael G\{u}nther; Maximilian Werk; Han Xiao; |
| 26 | TREC IKAT 2025: A Test Collection for The Offline and Interactive Evaluation of Conversational Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present the resources made for iKAT 2025, focusing on multi-session conversations (i.e., multiple dialogues per user), dynamically evolving user models, mixed-initiative dialogues, and large-scale human and automatic assessments. |
Zahra Abbasiantaeb; Simon Lupart; Marcel Gohsen; Nailia Mirzakhmedova; Johannes Kiesel; Jeffrey Dalton; Mohammad Aliannejadi; |
| 27 | Rank-R1: Enhancing Reasoning in LLM-based Document Rerankers Via Reinforcement Learning Related Papers Related Patents Related Grants Related Venues Related Experts Related Code View Save Highlight: We introduce Rank-R1, an LLM-based reranker that reasons over queries and candidate documents before ranking. |
Shengyao Zhuang; Xueguang Ma; Zheng Yao; Shuai Wang; Bevan Koopman; Jimmy Lin; Guido Zuccon; |
| 28 | Layer-wise Token Compression for Efficient Document Reranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose Layer-wise Token Compression (LTC), which applies adaptive token pooling at intermediate transformer layers. |
Shengyao Zhuang; Zhichao Xu; Ivano Lauriola; |
| 29 | Improving Interpretability of Cognitive Diagnosis Models with LLM-based Semantic Augmentation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As a result, providing interpretable diagnostic outputs becomes challenging. To address these limitations, we propose SACD, a semantic-augmented cognitive diagnosis framework that integrates LLM-based semantic analysis with student behavioral modeling. |
Youheng Bai; Jiaqi Zheng; Mingliang Hou; Teng Guo; Mi Tian; Xiangyu Zhao; Zitao Liu; Weiqi Luo; |
| 30 | Learning to Retrieve from Agent Trajectories Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we argue that retrieval models for agentic search should be trained directly from agent interaction data. |
Yuqi Zhou; Sunhao Dai; Changle Qu; Liang Pang; Jun Xu; Ji-Rong Wen; |
| 31 | GraphSynthQA: Knowledge-Graph-Guided Query Synthesis and Step-Level Preference Optimization for Web Agents Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, this task faces challenges with respect to data and training: existing QA datasets are mostly 1-3 hop over closed corpora (e.g., Wikipedia); meanwhile, outcome-based on-policy RL that used by recent works is inefficient and brittle in long-horizon, tool-heavy browsing environments. To address these challenges, we introduce GraphSynthQA, a knowledge-graph (KG)—guided synthesis framework in an open-web setting. |
Chiwei Zhu; Mingxuan Du; Benfeng Xu; Shengzhuo Zhang; Xiaorui Wang; Zhendong Mao; |
| 32 | Universal Item Tokenization for Transferable Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we propose UTGRec, a Universal item Tokenization approach for transferable Generative Recommendation. |
Bowen Zheng; Hongyu Lu; Yu Chen; Wayne Xin Zhao; Ji-Rong Wen; |
| 33 | Beyond The Single Path: Divergent Reasoning for LLM-based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paradigm is prone to reasoning path collapse, where limiting exploration of potentially superior and diverse reasoning paths within the LLMs space. As a result, both the accuracy and diversity of the recommendation outcomes are constrained.To address this issue, we propose a novel model, Divergent Reasoning for LLM-based Recommendation, named DivReason. |
Guojia An; Jie Zou; Yuhan Yang; Shuai Qin; Weikang Guo; Jinyu Guo; Yang Yang; |
| 34 | Task-Adaptive Retrieval Over Agentic Multi-Modal Web Histories Via Learned Graph Memory Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose ACGM, a learned graph-memory retriever that constructs task-adaptive relevance graphs over agent histories using policy-gradient optimization from downstream task success. |
Saman Forouzandeh; Kamal Berahmand; Mahdi Jalili; |
| 35 | APR: Adaptive Personalised Reranking For Conversational Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We show that instruction-following models can assist with complex queries but introduce noise and latency on simpler keyword queries. To address this issue, we propose Adaptive Personalised Reranking (APR), a framework that routes queries based on intent. |
Shen Dong; Iadh Ounis; Debasis Ganguly; |
| 36 | MVIGER: Multi-View Variational Integration of Complementary Knowledge for Generative Recommender Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Our in-depth quantitative analysis reveals that preference knowledge learned from diverse prompt templates and heterogeneous indices differs significantly, indicating a high potential for complementarity. To fully exploit this complementarity and provide consistent performance under varying prompts and item indices, we propose MVIGER, a unified variational framework that models selection among these information sources as a categorical latent variable with a learnable prior. |
Tongyoung Kim; SooJin Yoon; SeongKu Kang; Jinyoung Yeo; Dongha Lee; |
| 37 | HCPRA: A Hierarchical Cognition–Perception–Reasoning Agent Framework for Emotion-Cause Pair Extraction in Conversations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, in conversation scenarios, humans are the true origin of emotions, while language text merely serves as the medium of expression. Consequently, these methods lack the modeling of the cognition and reasoning processes behind the human’s behavior from the perspective of the speaker subject. |
Botao Wang; Lianwei Wu; Shuhan Guo; Kang Wang; Qingyan Wang; Tingran Zhang; Jiapeng Liu; Hikmat Ullah Khan; |
| 38 | WeatherArchive: A Benchmark for Retrieval-Augmented Reasoning Over Historical Weather Archives Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, their large scale, noise in optical character recognition (OCR), and archaic language make it difficult to transform them into structured knowledge for climate research. To address this challenge, we introduce \o{}urmethod, the first large-scale benchmark for evaluating end-to-end retrieval-augmented generation (RAG) systems on historical weather archives. |
Yongan Yu; Xianda Du; Qingchen Hu; Jiahao Liang; Jingwei Ni; Dan Qiang; Kaiyu Huang; Grant Mckenzie; Ren\'{e}e Sieber; Fengran Mo; |
| 39 | Are LLM-Based Retrievers Worth Their Cost? An Empirical Study of Efficiency, Robustness, and Reasoning Overhead Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We reproduce a reasoning-intensive retrieval benchmark (BRIGHT) across 12 tasks and 14 retrievers, and extend evaluation with cold-start indexing cost, query latency distributions and throughput, corpus scaling, robustness to controlled query perturbations, and confidence use (AUROC) for predicting query success. |
Abdelrahman Abdallah; Jamie Holdcroft; Mohammed Ali; Adam Jatowt; |
| 40 | Multi-Perspective Driven Expected Location Preferences for Next POI Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose a novel next POI recommendation method (MPDC) that leverages Multi-Perspective modeling and Diffusion-based Contrastive learning. |
Pengxiang Lan; Enneng Yang; Yuliang Liang; Jianzhe Zhao; Guibing Guo; Hai Zhao; |
| 41 | Well Begun Is Half Done: Training-Free and Model-Agnostic Semantically Guaranteed User Representation Initialization for Multimodal Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we propose a Semantically Guaranteed User Representation Initialization (SG-URInit). |
Jinfeng Xu; Zheyu Chen; Shuo Yang; Jinze Li; Hewei Wang; Jianheng Tang; Wei Wang; Xiping Hu; Edith C. H. Ngai; |
| 42 | Automating Generation of Long-Form Queries Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Traditional short keyword queries are increasingly being replaced by longer, more detailed queries that reflect complex and nuanced user information needs, especially in conversational assistants equipped with web search capabilities. In this work, we present a methodology for automatically generating such human-style long-form queries (narratives) by clustering raw short queries to form synthetic search sessions, designed to reflect a real user’s search behavior. |
Shivani Upadhyay; Daniel Campos; Nandan Thakur; Ronak Pradeep; Nick Craswell; Jimmy Lin; |
| 43 | When More Reformulations Hurt: Avoiding Drift Using Ranker Feedback Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we argue that the core challenge is not reformulation generation itself, but the adaptive selection of reformulations and their retrieved documents under a strict inference budget. |
Venktesh V; Mandeep Rathee; Avishek Anand; |
| 44 | Question-Adaptive Graph Learning for Multi-hop Retrieval Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Under the multi-hop settings, existing methods often struggle to fully understand the questions with complex semantic structures and are susceptible to irrelevant noise during the retrieval of multiple information targets. To address these limitations, we propose a novel graph representation learning framework for multi-hop question retrieval. |
Yuchen Yan; Peiyan Zhang; Zhihua Liu; Hao Wang; Yatao Bian; Weiming Li; Xiaoshuai Hao; |
| 45 | Deep Search with Hierarchical Meta-Cognitive Monitoring Inspired By Cognitive Neuroscience Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose Deep Search with Meta-Cognitive Monitoring (DS-MCM), a deep search framework augmented with an explicit hierarchical metacognitive monitoring mechanism. |
Zhongxiang Sun; Qipeng Wang; Weijie Yu; Jingxuan Yang; Haolang Lu; Jun Xu; |
| 46 | Towards End-to-End Alignment of User Satisfaction Via Questionnaire in Video Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Recently, explicit satisfaction feedback collected through questionnaires has emerged as a high-quality direct alignment supervision, but is extremely sparse and easily overwhelmed by abundant behavioral data, making it difficult to incorporate into online recommendation models. To address these challenges, we propose a novel framework which is towards End-to-End Alignment of user Satisfaction via Questionnaire, named EASQ, to enable real-time alignment of ranking models with true user satisfaction. |
Na Li; Jiaqi Yu; Minzhi Xie; Tiantian He; Xiaoxiao Xu; Zixiu Wang; Lantao Hu; Yongqi Liu; Han Li; Kaiqiao Zhan; Kun Gai; |
| 47 | Spectral Tempering for Embedding Compression in Dense Passage Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We show that the optimal scaling strength γ is not a global constant: it varies systematically with target dimensionality k and is governed by the signal-to-noise ratio (SNR) of the retained subspace. Based on this insight, we propose Spectral Tempering (SpecTemp), a learning-free method that derives an adaptive γ(k) directly from the corpus eigenspectrum using local SNR analysis and knee-point normalization, requiring no labeled data or validation-based search. |
Yongkang Li; Panagiotis Eustratiadis; Evangelos Kanoulas; |
| 48 | AlpsBench: An LLM Personalization Benchmark for Real-Dialogue Memorization and Preference Alignment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing benchmarks either overlook personalized information management that is critical for personalization or rely heavily on synthetic dialogues, which exhibit an inherent distribution gap from real-world dialogue. To bridge this gap, we introduce AlpsBench, An LLM PerSonalization benchmark derived from real-world human–LLM dialogues. |
Jianfei Xiao; Xiang Yu; Chengbing Wang; Wuqiang Zheng; Xinyu Lin; Kaining Liu; Hongxun Ding; Yang Zhang; Wenjie Wang; Fuli Feng; Xiangnan He; |
| 49 | Differentiable Semantic ID for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We attribute this to early deterministic assignments that limit codebook exploration, leading to imbalance and unstable optimization. In this paper, we therefore propose DIGER (Differentiable Semantic ID for GEnerative Recommendation). |
Junchen Fu; Xuri Ge; Alexandros Karatzoglou; Ioannis Arapakis; Suzan Verberne; Joemon M. Jose; Zhaochun Ren; |
| 50 | One Pass, Any Order: Position-Invariant Listwise Reranking for LLM-Based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Such order sensitivity makes LLM-based rerankers difficult to rely on, since rankings may reflect prompt serialization rather than user preference. We propose InvariRank, a permutation-invariant listwise reranking framework that addresses this dependence at the architectural level. |
Ethan Bito; Yongli Ren; Estrid He; |
| 51 | Multi-Vector Index Compression in Any Modality Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce four approaches for index compression, including a novel attention-guided clustering (\o{}urs). |
Hanxiang Qin; Alexander Martin; Rohan Jha; Chunsheng Zuo; Reno Kriz; Benjamin Van Durme; |
| 52 | StructAlign: Structured Cross-Modal Alignment for Continual Text-to-Video Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: A key challenge in CTVR is feature drift, which manifests in two forms: intra-modal feature drift caused by continual learning within each modality, and non-cooperative feature drift across modalities that leads to modality misalignment. To mitigate these issues, we propose StructAlign, a structured cross-modal alignment method for CTVR. |
Shaokun Wang; Weili Guan; Jizhou Han; Jianlong Wu; Yupeng Hu; Liqiang Nie; |
| 53 | Learning to Reason for Multi-Step Retrieval of Personal Context in Personalized Question Answering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose PR2 (Personalized Retrieval-Augmented Reasoning), a reinforcement learning framework that integrates reasoning and retrieval from personal context for personalization. |
Maryam Amirizaniani; Alireza Salemi; Hamed Zamani; |
| 54 | From Doxa to Logos in Scientific Peer Review Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Are reviewer claims grounded in the submitted paper? |
Negar Arabzadeh; Sajad Ebrahimi; Alireza Daghighfarsoodeh; Soroush Sadeghian; Seyed Mohammad Hosseini; Hai Son Le; Mahdi Bashari; Ebrahim Bagheri; |
| 55 | Bridging The Topology-Semantic Gap: A Benchmark and Framework for Power Grid Work Ticket Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We identify a fundamental Topological Association Modeling Deficiency in existing methods, manifesting as Retrieval Linkage Degradation (where semantic ambiguity severs links between abstract intents and specific technical entities) and Topological Reasoning Deficiency (failing to adhere to rigid physical interlocking rules, causing fatal safety violations). To address these bottlenecks, we introduce the Power Grid Work Ticket (PGWT) dataset to systematically evaluate topological reasoning under real-world noise and sparsity. |
Jiangbing Mao; Tianke Xiang; Yantong Zhu; Wenliang Liu; Qinggang Zhang; Zhihong Zhang; |
| 56 | BERT-Based Cross-Encoder for Large-Scale Engagement Prediction and Re-ranking in Walmart Search Engine Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Predicting such engagement preferences presents a more complex challenge than traditional relevance modeling, as it requires capturing nuanced query-item relationships that reflect both relevance and user intent. To capture these nuances, we extend semantic understanding to engagement prediction by learning directly from query-item text with engagement labels as supervision rather than relying on historical engagement statistics as input. |
Shuyi Chen; Philip Fu; Ajit Puthenputhussery; Changsung Kang; Ming Sun; Cun Mu; Sachin Yadav; Hongwei Shang; |
| 57 | Mine Over Yours: How Authorship Biases Evaluation in Generative Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We examine whether authorship biases evaluation—whether users judge AI output from their own interactions more favorably than equivalent output from others. |
Jeongwoo Ryu; Kyusik Kim; Soomin Kim; Jinsu Eun; Changhoon Oh; Bongwon Suh; |
| 58 | Mixture-of-Retrieval Experts for Reasoning-Guided Multimodal Knowledge Exploitation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing methods typically adhere to rigid retrieval paradigms by mimicking fixed retrieval trajectories and thus fail to fully exploit the knowledge of different retrieval experts through dynamic interaction based on the model’s knowledge needs or evolving reasoning states. To overcome this limitation, we introduce Mixture-of-Retrieval Experts (MoRE), a novel framework that enables MLLMs to collaboratively interact with diverse retrieval experts for more effective knowledge exploitation. |
Chunyi Peng; Zhipeng Xu; Zhenghao Liu; Yishan Li; Yukun Yan; Shuo Wang; Yu Gu; Minghe Yu; Ge Yu; Maosong Sun; |
| 59 | Balanced Co-Clustering of Users and Items for Embedding Table Compression in Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing solutions towards embedding compression either suffer from severely compromised recommendation accuracy or incur considerable computational costs. To mitigate these issues, this paper presents BACO, a fast and effective framework for compressing embedding tables. |
Runhao Jiang; Renchi Yang; Donghao Wu; |
| 60 | TRACE: A Conversational Framework for Sustainable Tourism Recommendation with Agentic Counterfactual Explanations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Traditional conversational travel recommender systems primarily optimize for user relevance and convenience, often reinforcing popular, overcrowded destinations and carbon-intensive travel choices. To address this, we present TRACE (Tourism Recommendation with Agentic Counterfactual Explanations), a multi-agent, LLM-based framework that promotes sustainable tourism through interactive nudging. |
Ashmi Banerjee; Adithi Satish; Wolfgang W\{o}rndl; Yashar Deldjoo; |
| 61 | FAVE: Flow-based Average Velocity Establishment for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While flow-based methods improve the efficiency of diffusion models, they remain hindered by the ”Noise-to-Data” paradigm, which introduces two critical inefficiencies: prior mismatch, where generation starts from uninformative noise, forcing a lengthy recovery trajectory; and linear redundancy, where iterative solvers waste computation on modeling deterministic preference transitions. To address these limitations, we propose a Flow-based Average Velocity Establishment (Fave) framework for one-step generation recommendation that learns a direct trajectory from an informative prior to the target distribution. |
Ke Shi; Yao Zhang; Feng Guo; Jinyuan Zhang; JunShuo Zhang; Shen Gao; Shuo Shang; |
| 62 | MCP Servers for Pyserini and RankLLM: Enabling Agentic Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present MCP servers for Pyserini and RankLLM, two widely used information retrieval research toolkits, enabling their retrieval and reranking capabilities to be easily accessed by LLM agents. |
Yijun Ge; Zibo Guo; Sahel Sharifymoghaddam; Jimmy Lin; |
| 63 | CARD: Non-Uniform Quantization of Visual Semantic Unit for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While existing studies have sought to enhance SID construction by incorporating multimodal content, collaborative signals, or more advanced quantization techniques, learning high-quality SIDs still faces two key challenges: (1) The two-stage generative recommendation paradigm (SID construction and autoregressive generation) provides insufficient supervision for heterogeneous fusion, which hinders learning high-quality SIDs, and (2) non-uniform embeddings lead to codeword imbalance and generation bias. To address these challenges, we propose a novel generative recommendation framework, called CARD. |
Yibiao Wei; Jie Zou; Pengfei Zhang; Xiao Ao; Weikang Guo; Zeyu Ma; Yang Yang; |
| 64 | GIGP+: A CPU-GPU Co-Processing Engine for Multi-Vector Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To get the best of both worlds, we propose GIGP+, a GPU-based method designed to achieve high parallelism and low computational overhead. |
Zheng Bian; Man Lung Yiu; Bo Tang; |
| 65 | GFlowGR: Fine-tuning Generative Recommendation Frameworks with Generative Flow Networks Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Although reward-based fine-tuning offers a partial remedy, it often lacks token-level supervision. To address these challenges, we reformulate GR as a sequential set-generation problem and propose GFlowGR, a GFlowNet-based fine-tuning framework that explicitly aligns generation probabilities with item-level utilities. |
Yejing Wang; Shengyu Zhou; Jinyu Lu; Qidong Liu; Xinhang Li; Wenlin Zhang; Feng Li; Pengjie Wang; Chuan Yu; Jian Xu; Bo Zheng; Xiangyu Zhao; |
| 66 | HyFormer: Revisiting The Roles of Sequence Modeling and Feature Interaction in CTR Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper presents HyFormer, a unified hybrid transformer architecture that tightly integrates long-sequence modeling and feature interaction into a single backbone. |
Yunwen Huang; Shiyong Hong; Xijun Xiao; Jinqiu Jin; Xuanyuan Luo; Zhe Wang; Zheng Chai; Shikang Wu; Yuchao Zheng; Jingjian Lin; |
| 67 | Learning to Summarize for Search Relevance with Reinforcement Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While full product descriptions provide richer information, their length and verbosity make them computationally impractical for real-time ranking, particularly when using cross-encoder architectures. To address this challenge, we propose ReLSum, a reinforcement learning framework that generates concise, relevance-optimized product summaries for search ranking. |
Nitin Yadav; Changsung Kang; Hongwei Shang; |
| 68 | Resources for Automated Evaluation of Assistive RAG Systems That Help Readers with News Trustworthiness Assessment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As the organizers of the DRAGUN track, we describe the resources that we have newly developed to allow for the reuse of the track’s tasks. |
Dake Zhang; Mark D. Smucker; Charles L. A. Clarke; |
| 69 | Optimal Re-Ranking Depth Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We formulate re-ranking depth as a query-specific property and study whether it can be predicted a priori from first-stage retrieval characteristics. Through a large-scale analysis, we find that most standard query performance prediction (QPP) methods are ineffective for this task. |
Siqing Huo; Andrew Parry; Debasis Ganguly; Charles L. A. Clarke; |
| 70 | Mining Informative Interests Via Latent Cross Reasoning for Search Enhanced Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Motivated by the human decision-making process, where one first identifies recommendation intent and then selectively reasons about relevant search signals, we propose LCR-SER, a latent cross reasoning method for search-enhanced recommendation. |
Teng Shi; Weicong Qin; Weijie Yu; Xiao Zhang; Ming He; Jianping Fan; Jun Xu; |
| 71 | Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, current systems predominantly operate on parsed text, which creates two critical bottlenecks: (1) Coarse-grained attribution, where users are burdened with manually locating evidence within lengthy documents based on vague text-level citations; and (2) Visual semantic loss, where the conversion of visually rich documents (e.g., slides, PDFs with charts) into text discards spatial logic and layout cues essential for reasoning. To bridge this gap, we present Chain of Evidence (CoE), a retriever-agnostic visual attribution framework that leverages Vision-Language Models to reason directly over screenshots of retrieved document candidates. |
Peiyang Liu; Ziqiang Cui; Xi Wang; Di Liang; Wei Ye; |
| 72 | Beyond Semantic Relevance: Counterfactual Risk Minimization for Robust Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In such cases, maximizing relevance paradoxically promotes the retrieval of sycophantic evidence that reinforces hallucinations, a critical failure we term the Relevance-Robustness Gap. To bridge this gap, we propose CoRM-RAG (Counterfactual Risk Minimization for RAG), a framework that aligns retrieval with decision safety rather than mere similarity. |
Peiyang Liu; Qiang Yan; Ziqiang Cui; Di Liang; Xi Wang; Wei Ye; |
| 73 | Multi-modal Relational Item Representation Learning for Inferring Substitutable and Complementary Items Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose MMSC, a self-supervised multi-modal relational representation learning framework that combines a multi-modal foundation model adapted to encode item metadata and a self-supervised denoising module that learns relationship-aware representations from noisy user behaviors, unified by a hierarchical aggregation mechanism. |
Junting Wang; Chenghuan Guo; Yang Jiao; Yanhui Guo; Hari Sundaram; Yan Gao; |
| 74 | When LLM Judges Inflate Scores: Exploring Overrating in Relevance Assessment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we conduct a study of overrating behavior in LLM-based relevance judgments across model backbones, evaluation paradigms (pointwise and pairwise), and passage modification strategies. |
Chuting Yu; Hang Li; Guido Zuccon; Joel Mackenzie; Teerapong Leelanupab; |
| 75 | Mitigating Adversarial Attacks By Transferring LLM-generated Narrative Reasoning for Robust Fake News Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Such attacks pollute both semantic and structural signals, causing GNN-based aggregators to fuse logically conflicting content and yield unreliable representations. To address this, we propose LLM-TKT, a novel framework that distills LLM-based narrative reasoning into lightweight GNNs for robust fake news detection. |
Mengyang Chen; Lingwei Wei; Wei Zhou; Songlin Hu; |
| 76 | Unifying Search and Recommendation in LLMs Via Gradient Multi-Subspace Tuning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Parameter-efficient fine-tuning (PEFT) offers a more practical alternative but faces two critical challenges in unifying S&R: (1) gradient conflicts across tasks due to divergent optimization objectives, and (2) shifts in user intent understanding caused by overfitting to fine-tuning data, which distort general-domain knowledge and weaken LLM reasoning. To address these issues, we propose Gradient Multi-Subspace Tuning (GEMS), a novel framework that unifies S&R with LLMs while alleviating gradient conflicts and preserving general-domain knowledge. |
Jujia Zhao; Zihan Wang; Shuaiqun Pan; Suzan Verberne; Zhaochun Ren; |
| 77 | Bridging The Gap: An End-to-End Framework for Decoupled Alignment in Dynamic Semantic ID Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Additionally, static generation fails to adapt to evolving data distributions, causing codebook drift. To address these limitations, we propose a dynamic End-to-End Semantic ID Generation Framework based on Decoupled Representation Alignment. |
Yu Cheng; Wei Xu; Li Li; Jianbin Lin; Can Ye; |
| 78 | Rethinking Semantic–Collaborative Integration: Why Alignment Is Not Enough Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose a complementary perspective in which semantic and collaborative representations are treated as partially shared yet fundamentally heterogeneous views, each containing both shared and view-specific factors. |
Maolin Wang; Dongze Wu; Jianing Zhou; Hongyu Chen; Beining Bao; Yu Jiang; Chenbin Zhang; Chang Wang; Jian Liu; Lei Sha; |
| 79 | Mitigating Collaborative Semantic ID Staleness in Generative Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, SID staleness under temporal drift is rarely analyzed explicitly. To bridge this gap, we study SID staleness under strict chronological evaluation and propose a lightweight, model-agnostic SID alignment update. |
Vladimir Baikalov; Iskander Bagautdinov; Sergey Muravyov; |
| 80 | Bi-Level Optimization for Generative Recommendation: Bridging Tokenization and Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This separation can lead to misalignment: the tokenizer is trained without direct guidance from the recommendation objective, potentially yielding suboptimal identifiers that degrade recommendation performance. To address this, we propose BLOGER, a Bi-Level Optimization for GEnerative Recommendation framework, which explicitly models the interdependence between the tokenizer and the recommender in a unified optimization process. |
Yimeng Bai; Chang Liu; Yang Zhang; Dingxian Wang; Frank Yang; Andrew Rabinovich; Wenge Rong; Fuli Feng; |
| 81 | Total Recall QA: A Verifiable Evaluation Suite for Deep Research Agents Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Inspired by prior research on TREC Total Recall Tracks, we introduce the task of Total Recall Question Answering and develop a framework for deep research agents evaluation that satisfies the identified criteria. |
Mahta Rafiee; Heydar Soudani; Zahra Abbasiantaeb; Mohammad Aliannejadi; Faegheh Hasibi; Hamed Zamani; |
| 82 | Topic-Specific Classifiers Are Better Relevance Judges Than Prompted LLMs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose to train topic-specific relevance classifiers instead: By finetuning monoT5 with independent LoRA weight adaptation on the judgments of a single assessor for a single topic’s pool, we align it to that assessor’s notion of relevance for that topic. |
Lukas Gienapp; Martin Potthast; Andrew Yates; Harrisen Scells; Eugene Yang; |
| 83 | Understanding DNNs in Feature Interaction Models: A Dimensional Collapse Perspective Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present a novel perspective to understand the effectiveness of DNNs: their impact on the dimensional robustness of the representations. |
Jiancheng Wang; Mingjia Yin; Hao Wang; Enhong Chen; |
| 84 | A Sensitivity-Aware Test Collection for Search Among Personal Information Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present baseline performances for relevance, sensitivity classification, and sensitivity-aware search on the collection. |
Jack McKechnie; Graham McDonald; Craig Macdonald; |
| 85 | Synthetic Data Powers Product Retrieval for Long-tail Knowledge-Intensive Queries in E-commerce Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: These queries exhibit diverse linguistic patterns, often lack explicit purchase intent, and require domain-specific knowledge reasoning for accurate interpretation. They also suffer from a shortage of reliable behavioral logs, which makes such queries a persistent challenge for retrieval optimization.To address these issues, we propose an efficient data synthesis framework tailored to retrieval involving long-tail, knowledge-intensive queries. |
Gui Ling; Weiyuan Li; Yue Jiang; Wenjun Peng; Xingxian Liu; Dongshuai Li; Fuyu Lv; Dan Ou; Haihong Tang; |
| 86 | AsarRec: Adaptive Sequential Augmentation for Robust Self-supervised Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, these methods heavily rely on their pre-defined static augmentation strategies~(where the augmentation type remains fixed once chosen) to construct augmented views, leading to two critical challenges: (1) the optimal augmentation type can vary significantly across different scenarios; (2) inappropriate augmentations may even degrade recommendation performance, limiting the effectiveness of SSL. To overcome these limitations, we propose an adaptive augmentation framework. |
Kaike Zhang; Qi Cao; Fei Sun; Xinran Liu; Huawei Shen; Xueqi Cheng; |
| 87 | CSyMR: Benchmarking Compositional Music Information Retrieval in Symbolic Music Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce CSyMR-Bench, a benchmark for compositional MIR in symbolic music reasoning grounded in authentic user scenarios. |
Boyang Wang; Yash Vishe; Xin Xu; Zachary Novack; Xunyi Jiang; Julian McAuley; Junda Wu; |
| 88 | NanoKnow: How to Know What Your Language Model Knows Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To demonstrate NanoKnow’s utility, we conduct experiments using eight nanochat checkpoints. |
Lingwei Gu; Nour Jedidi; Jimmy Lin; |
| 89 | Revisiting BM25 Feedback Models Using HyDE Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we revisit and systematically evaluate traditional BM25 feedback models in the context of HyDE, a popular method that enriches query representations using LLM-generated hypothetical answer documents. |
Nour Jedidi; Jimmy Lin; |
| 90 | LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval Via A Two-Phase Training Curriculum Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we introduce LACONIC, a family of learned sparse retrievers based on the Llama3 architecture (1B, 3B, and 8B). |
Zhichao Xu; Shengyao Zhuang; Crystina Zhang; Xueguang Ma; Yijun Tian; Maitrey Mehta; Jimmy Lin; Vivek Srikumar; |
| 91 | Minimal-Perturbation Counterfactuals Through Guided Denoising Diffusion for Recommender Systems Explanation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose DiceRec, a diffusion-based, model-agnostic framework for generating interaction-level counterfactual explanations. |
Amir Reza Mohammadi; Andreas Peintner; Michael M. M\{u}ller; Eva Zangerle; |
| 92 | Good Ranks Follow Good Answers: Unsupervised Answer-Driven Reranking for Multimodal Document QA Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing reranker training frameworks in MDQA rely predominantly on proxy supervision derived from human annotations or large language model (LLM) outputs, which are frequently noisy and, more critically, misaligned with downstream answer quality. To overcome this limitation, we propose AD-Reranker, a novel framework that shifts reranker training from proxy imitation to answer-driven utility optimization. |
Keyu Zhu; Shuanghong Shen; Xianquan Wang; Kai Zhang; Shijin Wang; Qi Liu; Zhenya Huang; |
| 93 | Decoding Multimodal Cues: Unveiling The Implicit Meaning Behind Hateful Videos Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing studies primarily focus on binary classification and fail to provide contextual rationales that reveal the implicit meanings behind these judgments, significantly undermining model explainability. To fill this gap, we aim to achieve explainable hateful video detection, enabling models to provide contextual rationales that integrate relevant evidence and logical reasoning alongside decisions. |
Junyu Lu; Deyi Ji; Liqun Liu; Xiaokun Zhang; Youlin Wu; Roy Ka-Wei Lee; Peng Shu; Huan Yu; Jie Jiang; Bo Xu; Liang Yang; Hongfei Lin; |
| 94 | Robust Multimodal Recommendation Via Graph Retrieval-Enhanced Modality Completion Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, these methods may overlook semantically relevant context in the graph, which contains valuable cues that are non-trivial to capture through simple methods like neighborhood aggregation. In this work, we propose GRE-MC, a Graph Retrieval–Enhanced Modality Completion framework, to overcome these limitations. |
Yuan Li; Jun Hu; Jiaxin Jiang; Bryan Hooi; Bingsheng He; |
| 95 | Understanding Wacky Weights: A Dissection of SPLADE’s Learned Term Importance Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we reproduce SPLADE-v2 to systematically investigate wacky weights across the SPLADE family of models. |
Gregory Polyakov; Harrisen Scells; Carsten Eickhoff; |
| 96 | CoverageBench: Evaluating Information Coverage Across Tasks and Domains Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we construct a benchmark, CoverageBench, for evaluating information coverage made from existing collections. |
Saron Samuel; Andrew Yates; Dawn Lawrie; Ian Soboroff; Trevor Adriaanse; Benjamin Van Durme; Eugene Yang; |
| 97 | QE-RAG: A Robust Retrieval-Augmented Generation Benchmark for Query Entry Errors Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We analyze the impact of these errors on LLM outputs and find that corrupted queries degrade model performance, which can be mitigated through query correction and training a robust retriever for retrieving relevant documents. Based on these insights, we propose a contrastive learning-based robust retriever training method and a retrieval-augmented query correction method. |
Kepu Zhang; Zhongxiang Sun; Weijie Yu; Xiaoxue Zang; Kai Zheng; Yang Song; Han Li; Jun Xu; |
| 98 | Conv-FinRe: A Conversational and Longitudinal Benchmark for Utility-Grounded Financial Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce Conv-FinRe, a conversational and longitudinal benchmark for stock recommendation that evaluates LLMs beyond behavior matching. |
Yan Wang; Yi Han; Lingfei Qian; Yueru He; Xueqing Peng; Dongji Feng; Zhuohan Xie; Vincent Jim Zhang; Yuqing Guo; Fengran Mo; Jimin Huang; Yankai Chen; Jian-Yun Nie; |
| 99 | FinAuditing: A Financial Taxonomy-Structured Multi-Document Benchmark for Evaluating LLMs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce FinAuditing, a taxonomy-aligned, structure-aware benchmark built from real XBRL filings. |
Yan Wang; Keyi Wang; Shanshan Yang; Jaisal Patel; Jeff Zhao; Fengran Mo; Xueqing Peng; Lingfei Qian; Yankai Chen; V\'{\i}ctor Guti\'{e}rrez-Basulto; Jimin Huang; Guojun Xiong; Xiao-Yang Liu; Jian-Yun Nie; |
| 100 | Benchmarking Filtered Approximate Nearest Neighbor Search Algorithms on Transformer-based Embedding Vectors Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: By performing an in-depth literature analysis on FANNS, we identify a key gap in the research landscape: publicly available datasets with embedding vectors from state-of-the-art transformer-based text embedding models that contain abundant real-world attributes covering a broad spectrum of attribute types and value distributions. To fill this gap, we introduce the arxiv-for-fanns dataset of transformer-based embedding vectors for the abstracts of over 2.7 million arXiv papers, enriched with 11 real-world attributes such as authors and categories. |
Patrick Iff; Paul Br\{u}gger; Marcin Chrapek; David Kochergin; Maciej Besta; Torsten Hoefler; |
| 101 | Beyond Single Slot: Joint Optimization for Multi-Slot Guaranteed Display Advertising Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose a novel joint optimization framework for multi-slot GD allocation, addressing key challenges such as slot-level redundancy, contract imbalance, and exposure concentration. |
Zhaoqi Zhang; Jiaming Deng; Miao Xie; Linyou Cai; Qianlong Xie; Xingxing Wang; Siqiang Luo; Gao Cong; |
| 102 | CoCo: Conformal Confidence Suppression to Optimize Search Results Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose a robust content suppression mechanism to selectively suppress content when necessary. |
Zhou Qin; Shuang Wu; Yoon Kim; Yi Liu; Wenyang Liu; |
| 103 | MLLMRec: A Preference Reasoning Paradigm with Graph Refinement for Multimodal Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing methods still encounter two key problems in the representation learning of users and items, respectively: (1) the initialization of multimodal user representations is either agnostic to historical behaviors or contaminated by irrelevant modal noise, and (2) the widely used KNN-based item-item graph contains noisy edges with low similarities and lacks audience co-occurrence relationships. To address such issues, we propose MLLMRec, a novel preference reasoning paradigm with graph refinement for multimodal recommendation. |
Yuzhuo Dang; Xin Zhang; Zhiqiang Pan; Yuxiao Duan; Wanyu Chen; Fei Cai; Honghui Chen; |
| 104 | ProEchoMem: Enhancing Long Video Understanding Via Multi-Trace Probe-Echo Memory Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Inspired by Multiple-Trace Theory in cognitive psychology, we revisit long video understanding from a probe-echo perspective, in which human episodic memories are activated and integrated in parallel. Building on this insight, we propose ProEchoMem, a cognitive-inspired framework that simulates the probe-echo mechanism: (1) Incremental Episodic Memory Construction builds structured knowledge graphs from video streams; (2) Probe-Driven Memory Activation generates probe signals from user queries to activate all stored traces simultaneously; (3) Memory Echo Synthesis integrates activated traces into a coherent and structured memory echo. |
Derong Xu; Yanxin Chen; Wanyu Wang; Pengyue Jia; Chao Zhang; Maolin Wang; Yiqi Wang; Jipeng Qiang; Xuetao Wei; Hongzhi Yin; Tong Xu; Xiangyu Zhao; |
| 105 | Enhancing Enterprise Assistant Responses with Rich Multimodal Artifacts from Product Documentation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Modern enterprise documentation is rich with images (screenshots, flowcharts, system diagrams, etc.) and instructional videos (product tutorials), yet typical conversational systems often provide text-only answers that fail to capture procedural or interface-level nuances. We present our in-production retrieval framework designed to enrich conversational responses by effectively surfacing these multimodal artifacts. |
Sohan Patnaik; Sai Sree Harsha; Kowndinya Renduchintala; Milan Aggarwal; Sumit Bhatia; Yunyao Li; |
| 106 | RULER: Robust Unified LLM-based Efficient Retrieval for Legal Information Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Legal information retrieval demands high precision, yet traditional ”Retrieve-then-Rerank” pipelines with two separate models suffer from cascading error propagation and knowledge disconnects between stages. To address these issues, we propose RULER, a Robust Unified LLM-based Efficient Retrieval that integrates efficient Bi-Encoder retrieval and high-precision Cross-Encoder reranking within a parameter-sharing architecture. |
Chenyu Hou; Ziyang Wang; Bin Cao; Jiaxing Wang; Tianming Zhang; Tiantian Li; |
| 107 | Verifiable Reasoning for LLM-based Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, this paradigm inevitably suffers from reasoning degradation (i.e., homogeneous or error-accumulated reasoning) due to the lack of intermediate verification, thus undermining the recommendation. To bridge this gap, we propose a novel reason-verify-recommend paradigm, which interleaves reasoning with verification to provide reliable feedback, guiding the reasoning process toward more faithful user preference understanding. |
Xinyu Lin; Hanqing Zeng; Hanchao Yu; Yinglong Xia; Jiang Zhang; Aashu Singh; Fei Liu; Wenjie Wang; Fuli Feng; Tat-Seng Chua; Qifan Wang; |
| 108 | Towards A General Intelligent Information Agent: Framework, Models, and Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose to formalize GIANT generally as an agent interacting with its users to minimize their effort on finishing a task as well as the agent’s resource overhead. We propose a probabilistic modeling framework for optimizing GIANT’s interactions with its users and discuss how to estimate its four component models, including Situation Model, Content Model, Task Model, and User Model. |
Chengxiang Zhai; |
| 109 | Individual Turing Test: A Case Study of LLM-based Simulation Using Longitudinal Personal Data Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper presents a case study investigating LLM-based individual simulation using a volunteer-contributed archive of private messaging history spanning over ten years. |
Minghao Guo; Ziyi Ye; Wujiang Xu; Xi Zhu; Wenyue Hua; Dimitris N. Metaxas; |
| 110 | PLAID-PRF: Pseudo-Relevance Feedback with Centroid-like Tokens in PLAID Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we introduce PLAID-PRF, a method that performs Pseudo-Relevance Feedback (PRF) over PLAID to reformulate ColBERT’s query vectors based on the top-retrieved results. |
Xiao Wang; Sean MacAvaney; Craig Macdonald; |
| 111 | ColBERTSaR: Sparsified ColBERT Index Via Product Quantization Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose an embedding quantization approach that turns a ColBERT index into a true inverted index. |
Eugene Yang; Andrew Yates; Dawn Lawrie; James Mayfield; Saron Samuel; Rohan Jha; |
| 112 | Reproducing Adaptive Reranking for Reasoning-Intensive IR Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We observe that the quality of the reranker’s signal plays an important role in identifying additional relevant documents within the corpus graph. |
Mandeep Rathee; Venktesh V; Sean MacAvaney; Avishek Anand; |
| 113 | Think with Grounding: Curriculum Reinforced Reasoning with Video Grounding for Long Video Understanding Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, the existing literature suffers from the fact that the text-only reasoning under fixed video context may exacerbate hallucinations since detailed crucial clues are often ignored under limited video context length due to the temporal redundancy of long videos. To address this gap, we propose Video-TwG, a curriculum reinforced framework that employs a novel Think-with-Grounding paradigm, enabling video LLMs to actively decide when to perform on-demand grounding during interleaved text–video reasoning, selectively zooming into question-relevant clips only when necessary. |
Houlun Chen; Xin Wang; Guangyao Li; Yuwei Zhou; Yihan Chen; Jia Jia; Wenwu Zhu; |
| 114 | When and How to Ask: Dynamic Preference Elicitation Strategies for Conversational Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we conduct a systematic investigation of preference elicitation strategies from a stage-aware perspective. |
Feng Xia; Shuo Zhang; Xi Wang; |
| 115 | Sim.API: A Middleware to Simplify The Use of User Simulators for Shared Tasks in Conversational Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present Sim.API, a middleware-based run submission system that exposes user simulators through simple API endpoints and records run information without having to rely on run files. |
Marcel Gohsen; Nailia Mirzakhmedova; Zahra Abbasiantaeb; Johannes Kiesel; Simon Lupart; Jeffrey Dalton; Benno Stein; Mohammad Aliannejadi; |
| 116 | Does LLM Relevance Labelling Work for Arabic? Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Large Language Models (LLMs) are increasingly used in Information Retrieval, both within retrieval pipelines and for constructing evaluation resources. |
Marwah Alaofi; Fatima Haouari; |
| 117 | AdversarialCoT: Single-Document Retrieval Poisoning for LLM Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Unlike prior work that floods the corpus with poisoned documents, we propose AdversarialCoT, a query-specific attack that poisons only a single document in the corpus. |
Hongru Song; Yu-An Liu; Ruqing Zhang; Jiafeng Guo; Maarten de Rijke; Yixing Fan; Xueqi Cheng; |
| 118 | FACE: A Fine-Grained Reference-Free Evaluator for Conversational Information Access Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose FACE: a Fine-grained, Aspect-based Conversation Evaluator that evaluates diverse turn- and dialogue-level aspects of conversations. |
Hideaki Joko; Faegheh Hasibi; |
| 119 | A Replicability Study of Joint Product Quantisation for Effective Space-Efficient Dense Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Joint Product Quantisation (JPQ) improves over PQ by further training the obtained centroid embeddings, as well as the query encoder, for enhanced effectiveness. In this replicability study, we reimplement JPQ, comparing its performance on the STAR biencoder model with that reported in the original paper, as well as considering its generalisability in several dimensions: to other biencoders such as TCT and TAS-B; to larger models such as RepLLama; and beyond document retrieval, to a retrieval augmented generation task. |
Craig Macdonald; Zhlli Shen; Nicola Tonellotto; |
| 120 | Towards Knowledgeable Deep Research: Framework and Benchmark Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In contrast to existing studies that mainly focus on unstructured web content, a more challenging DR task should additionally utilize structured knowledge to provide a solid data foundation, facilitate quantitative computation, and lead to in-depth analyses. In this paper, we refer to this novel task as Knowledgeable Deep Research (KDR), which requires DR agents to generate reports with both structured and unstructured knowledge. |
Wenxuan Liu; Zixuan Li; Long Bai; Chunmao Zhang; Fenghui Zhang; Zhuo Chen; Wei Li; Yuxin Zuo; Fei Wang; Bingbing Xu; Xuhui Jiang; Jin Zhang; Xiaolong Jin; Jiafeng Guo; Tat-Seng Chua; Xueqi Cheng; |
| 121 | R3Check: Reinforcement Learning for Iterative Retrieval and Structured Reasoning in Complex Fact Checking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose R3Check, a rule-guided reinforcement learning framework that enables LLMs to perform iterative retrieval–reasoning for multi-hop fact-checking. |
Peng Qi; Yuyang Zhao; Wynne Hsu; Mong Li Lee; |
| 122 | Is A Busy Search Agent A Good One? Overthinking and Overretrieval at Scale Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we conduct a systematic study of overthinking and overretrieval in search agents from a scaling perspective. |
Xin Liu; Ruqing Zhang; Yu-An Liu; Lixin Su; Jiafeng Guo; Xueqi Cheng; |
| 123 | Incentivizing Retrieval-Augmented Generation Via Inner Adaptive Context Selection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we introduce InnerRAG, which incentivizes RAG via Inner Adaptive Context Selection. |
Chenxu Cui; Lin Shen; Haihui Fan; Sa Zhu; Feifei Dai; Bo Li; |
| 124 | DIVER: Unlocking Diversity in Ad Headline Generation with Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Meanwhile, existing diversity-enhancing techniques like stochastic decoding frequently compromise semantic coherence and controllability. To break this trade-off, we propose DIVER, an automated training framework that internalizes diversity as an intrinsic model capability. |
Chang Wang; Siyu Yan; Depeng Yuan; Yuqi Chen; Yanhua Huang; Yuanhang Zheng; Shuhao Li; Yinqi Zhang; Kedi Chen; Mingrui Zhu; Ruiwen Xu; |
| 125 | Learning to Route Queries to Heads for Attention-based Re-ranking with Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose a query-dependent head selection method, RouteHead, for attention-based re-ranking with LLMs. |
Yuxing Tian; Fengran Mo; Zhiqi Huang; Weixu Zhang; Jian-Yun Nie; |
| 126 | Scaling and Stabilizing Large-Scale Embedding-Based Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present a unified pipeline deployed at Walmart that addresses both signal quality and model evolution. |
Zhen Yang; Juexin Lin; Hongwei Shang; Kaihao Li; Feng Liu; Satya Chembolu; Xunfan Cai; Xinyi Liu; Cun Mu; Tony Lee; Ciya Liao; |
| 127 | Full Retraining, Incremental Fine-tuning, and Hybrid Serving: Model Updating and Serving for Industrial Generative Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Under certain practically relevant conditions, the proposed hybrid serving framework provides a useful balance between adaptation, stability, and efficiency. |
Lu Fan; Qijiong Liu; Zhongzhou Liu; Guoyuan An; Wei Guo; Yong Liu; Xiao-Ming Wu; |
| 128 | LLM-Oriented Information Retrieval: A Denoising-First Perspective Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this perspective paper, we argue that denoising-maximizing usable evidence density and verifiability within a context window-is becoming the primary bottleneck across the full information access pipeline. |
Lu Dai; Liang Sun; Fanpu Cao; Ziyang Rao; Cehao Yang; Hao Liu; Hui Xiong; |
| 129 | Fourier Kolmogorov-Arnold Network and Hypergraph Enhanced Contrastive Learning for Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Although self-supervised learning methods have been introduced to address these issues, these methods often overlook the intricate dependencies between users and items and fail to effectively utilize high-order global information. To address these challenges, we propose Fourier Kolmogorov-Arnold Network and Hypergraph Enhanced Contrastive Learning (FHCL) for recommendation. |
Yuwen Liu; Lianyong Qi; Xucheng Zhou; Xingyuan Mao; Weiming Liu; Shuang Wang; Xiaolong Xu; Haolong Xiang; Xuyun Zhang; Wanchun Dou; |
| 130 | A Graph-Enhanced MLLM for Hierarchical Multimodal Emotion Understanding and Support in Conversations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Moreover, their multimodal fusion relies on coarse-grained concatenation of encoder features into the prompt, ignoring structured relational dependencies among speakers and modalities. To address these limitations, we propose GraphEMO, a unified multimodal emotion understanding and support framework based on a graph-enhanced MLLM. |
Geng Tu; Taiyu Niu; Xi Zeng; Ruifeng Xu; Min Zhang; |
| 131 | Agentic Search in The Wild: Intents and Trajectory Dynamics from 14M+ Real Search Requests Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper presents a large-scale log analysis of agentic search based on 14.44M search requests (3.97M sessions) collected from DeepResearchGym, i.e., an open-source search API accessed by external agentic clients. |
Jingjie Ning; Jo\~{a}o Coelho; Yibo Kong; Yunfan Long; Bruno Martins; Jo\~{a}o Magalh\~{a}es; Jamie Callan; Chenyan Xiong; |
| 132 | Learning Noise-Resilient and Transferable Graph-Text Alignment Via Dynamic Quality Assessment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: These limitations reveal a fundamental trade-off: leveraging expressive many-to-many signals increases semantic coverage but may propagate errors under noise, whereas strict one-to-one training is more conservative yet still suffers when mismatched pairs remain in the training set. Therefore, we propose ADAligner, a quality-aware graph–text alignment framework that adapts between expressive many-to-many and conservative one-to-one objectives based on estimated alignment reliability. |
Yuhang Liu; Minglai Shao; Zengyi Wo; Yunlong Chu; Bing Hao; Shengzhong Liu; Ruijie Wang; Jianxin Li; |
| 133 | BEAR: Towards Beam-Search-Aware Optimization for Recommendation with Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, we identify a critical training-inference inconsistency: while SFT optimizes the overall probability of positive items, it does not guarantee that such items will be retrieved by beam search even if they possess high overall probabilities. Due to the greedy pruning mechanism, beam search can prematurely discard a positive item once its prefix probability is insufficient.To address this inconsistency, we propose BEAR (Beam-SEarch-Aware Regularization), a novel fine-tuning objective that explicitly accounts for beam search behavior during training. |
Weiqin Yang; Bohao Wang; Zhenxiang Xu; Jiawei Chen; Shengjia Zhang; Jingbang Chen; Canghong Jin; Can Wang; |
| 134 | Efficient Memory Alignment for Long-term Conversational Information Seeking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We frame persona memory as a retrieval problem over a growing memory store, and propose REMAP, a reflection-guided memory editing approach for online alignment of persona facts that selectively writes and revises memory entries based on the current dialogue evidence and retrieved related items. |
Qingyang Xu; Xiao Liu; Zhouhua Fang; Yong Li; Zhiwei Liu; Vincent Lee; Haishuai Wang; |
| 135 | KARMA: Knowledge-Action Regularized Multimodal Alignment for Personalized Search at Taobao Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose KARMA (Knowledge–Action Regularized Multimodal Alignment), a unified framework that treats semantic reconstruction as a train-only regularizer. |
Zhi Sun; Wenming Zhang; Yi Wei; Liren Yu; Zhixuan Zhang; Dan Ou; Haihong Tang; |
| 136 | Humans, LLMs, and Measures Do Not Align in Attributed Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Abstract: Evaluating attributed information retrieval (AIR) systems requires assessing both informativeness and attributability. To enable scalable evaluation, LLM-sourced ground truth data … |
Lukas Gienapp; Jenny Lang; Martin Potthast; Harrisen Scells; |
| 137 | Scaling Laws for Embedding Dimension in Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we conduct a comprehensive analysis of the relationship between embedding dimension and retrieval performance. |
Julian Killingback; Mahta Rafiee; Madine Manas; Hamed Zamani; |
| 138 | Personalized Deep Research: A User-Centric Framework, Dataset, and Hybrid Evaluation for Knowledge Discovery Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Current systems fail to adaptively adjust the depth and breadth of exploration based on the user’s existing expertise or latent interests, frequently resulting in reports that are either redundant for experts or overly dense for novices. To address this, we introduce Personalized Deep Research (PDR), a framework that integrates dynamic user context into the core retrieval-reasoning loop. |
Xiaopeng Li; Wenlin Zhang; Yingyi Zhang; Pengyue Jia; Yejing Wang; Yichao Wang; Yong Liu; Huifeng Guo; Xiangyu Zhao; |
| 139 | Comparing Token Pruning Approaches for Multi-Vector Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We compare standard pruning approaches like weighted token pruning or IDF-based pruning and analyze the impact on the downstream effectiveness of respective ColBERT models. |
Ferdinand Schlatt; Hanno Barschel; Matthias Hagen; |
| 140 | Towards A Relevance Posterior in Neural Information Access Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present empirical evidence that incorporating query-independent document utility can complement existing rankers and improve effectiveness with minimal query-time computation (solely score fusion). |
Andrew Parry; Emmanouil Georgios Lionis; Debasis Ganguly; Sean MacAvaney; |
| 141 | Why Advanced Encoders Lag on Sparse Retrieval? The Answer and An Approach to Bridging Vocabulary Gaps Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We formalize this intuition through a theoretical framework, demonstrating that appropriate vocabulary coarse-graining can tighten the generalization bounds by reducing complexity of the hypothesis class, provided that semantic integrity is preserved. To resolve this, we propose Vocabulary Transfer (VT), a model-agnostic framework that migrates advanced encoders to sparse-friendly, normalized vocabularies with minimal computational cost. |
Zhichao Geng; Yang Yang; |
| 142 | TEC: A Collection of Human Trial-and-error Trajectories for Problem Solving Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: The core issue is the absence of appropriate data: current models cannot learn from detailed records of how humans actually conduct trial-and-error in practice. To address this gap, we introduce a data annotation platform and a corresponding dataset, termed Trial-and-Error Collection (TEC). |
Xinkai Zhang; Jingtao Zhan; Yiqun Liu; Qingyao Ai; |
| 143 | Equip Pre-ranking with Target Attention By Residual Quantization Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This disparity creates a significant performance bottleneck for the entire system. To bridge this gap, we propose TARQ, a novel pre-ranking framework. |
Yutong Li; Yu Zhu; Yichen Qiao; Ziyu Guan; Lv Shao; Tong Liu; Bo Zheng; |
| 144 | Search for Coverage: Learning Coverage-Aware Retrieval with Augmented Sub-Question Answerability Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose CoveR, a dense retrieval method optimized for coverage-aware retrieval scenarios. |
Jia-Huei Ju; Eugene Yang; Trevor Adriaanse; Suzan Verberne; Andrew Yates; |
| 145 | Deletion Isn’t Enough: Auditing RAG for Selective Forgetting Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper highlights a research gap: non-disclosure obligations often concern propositions that must not be stated, while deployed Retrieval-Augmented Generation (RAG) systems enforce restrictions through record-level handling such as document removal or access-control lists. |
Leila Tavakoli; Mark Sanderson; |
| 146 | Pay Attention to Sequence Split: Uncovering The Impacts of Sub-Sequence Splitting on Sequential Recommendation Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: SSS demonstrates strong capabilities only when specific splitting methods, target strategies, and loss functions are used together. |
Yizhou Dang; Yifan Wu; Minhan Huang; Chuang Zhao; Lianbo Ma; Guibing Guo; Xingwei Wang; Zhu Sun; |
| 147 | STRIDE: Strategic Iterative Decision-Making for Retrieval-Augmented Multi-Hop Question Answering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: On the other hand, existing methods overlook the logical dependencies among reasoning steps, resulting in uncoordinated execution. To address these issues, we propose STRIDE, a framework that separates strategic planning, dynamic control, and grounded execution. |
Wei Chen; Lili Zhao; Zhi Zheng; Huijun Hou; Tong Xu; |
| 148 | DeepResearch-9K: A Challenging Benchmark Dataset of Deep-Research Agent Related Papers Related Patents Related Grants Related Venues Related Experts View Save Abstract: Deep-research agents are capable of executing multi-step web exploration, targeted retrieval, and sophisticated question answering. Despite their powerful capabilities, … |
Tongzhou Wu; Yuhao Wang; Xinyu Ma; Xiuqiang He; Shuaiqiang Wang; Dawei Yin; Xiangyu Zhao; |
| 149 | CASE: Cadence-Aware Set Encoding for Large-Scale Next Basket Repurchase Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present CASE (Cadence-Aware Set Encoding) for next basket repurchase recommendation, which decouples item-level cadence learning from cross-item interaction, enabling explicit calendar-time modeling while remaining production-scalable. |
Yanan Cao; Ashish Ranjan; Sinduja Subramaniam; Evren Korpeoglu; Kaushiki Nag; Kannan Achan; |
| 150 | BioCLEAR Benchmark for Biomedical Text Simplification Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Access to objective, reliable scientific information is crucial in a world of misinformation and disinformation, yet the general public often avoids scientific literature due to its perceived complexity. Modern generative information access models hold the promise of removing some of these barriers by developing appropriate approaches to scientific text simplification. |
Jan Bakker; Liana Ermakova; Jaap Kamps; |
| 151 | TimeMM: Time-as-Operator Spectral Filtering for Dynamic Multimodal Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Despite strong progress, most multimodal recommenders still rely on static interaction graphs or coarse temporal heuristics, which limits their ability to model continuous preference evolution with fine-grained temporal adaptation. To address these limitations, we propose TimeMM, a time-conditioned spectral filtering framework for dynamic multimodal recommendation. |
Wei Yang; Rui Zhong; Zihan Lin; Xiaodan Wang; Cheng Chen; Huan Ren; Yao Hu; |
| 152 | Visual-RAG: Benchmarking Text-to-Image Retrieval Augmented Generation for Visual Knowledge Intensive Queries Related Papers Related Patents Related Grants Related Venues Related Experts Related Code View Save Highlight: We introduce Visual-RAG, a question-answering benchmark that targets visually-grounded, knowledge-intensive questions in a visual evidence-centric manner. |
Yin Wu; Quanyu Long; Jing Li; Jianfei Yu; Wenya Wang; |
| 153 | EVADE-Bench: Multimodal Benchmark for Evaluating and Enhancing Evasive Content Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This gap is particularly consequential in e-commerce, where accurate moderation demands that both capabilities operate in concert. To address this gap, we introduce EVADE-Bench, the first expert-curated Chinese multimodal benchmark specifically designed to evaluate LLMs and VLMs on evasive content detection in real-world e-commerce scenarios. |
Ancheng Xu; Zhihao Yang; Jingpeng Li; Guanghu Yuan; Longze Chen; Liang Yan; Jiehui Zhou; Zhen Qin; Hengyu Chang; Yukun Chen; Hamid Alinejad-Rokny; Min Yang; |
| 154 | Disentangling from Collaborative and Semantic Views: Graph Collaborative Filtering for Q&A Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Traditional recommendation methods treat the question-answer pair as a whole or only consider the answer as a single item, which overlooks the two challenges and cannot effectively model user interests. To address these challenges, we introduce a graph neural network model named Question & Answer Graph Collaborative Filtering (QAGCF). |
Changshuo Zhang; Teng Shi; Xiao Zhang; Yanping Zheng; Ruobing Xie; Qi Liu; Jun Xu; |
| 155 | RoTE: Coarse-to-Fine Multi-Level Rotary Time Embedding for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While effective, such a process only considers the order of temporal information but overlooks the actual time spans between interactions, resulting in a coarse representation of users’ temporal dynamics and limiting the model’s ability to capture long-term and short-term interest evolution. To address this limitation, we propose RoTE, a novel multi-level temporal embedding module that explicitly models time span information in sequential recommendation. |
Haolin Zhang; Longtao Xiao; Guohao Cai; Ruixuan Li; Xiu Li; |
| 156 | Better Than Dense? Investigating The Natural Backward Compatibility of Learned Sparse Representations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We explore lightweight query adaptation methods including ranking fusion, representation fusion, and minimal-training adapters to further improve compatibility. |
Jingfen Qiao; Gabrielle Poerwawinata; Thong Nguyen; Jia-Huei Ju; Eugene Yang; Evangelos Kanoulas; Andrew Yates; |
| 157 | Exploring Knowledge Conflicts for Faithful LLM Reasoning: Benchmark and Method Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Meanwhile, modern RAG systems increasingly emphasize the integration of unstructured text and (semi-)structured data like knowledge graphs (KGs) to improve knowledge completeness and reasoning faithfulness. To address this gap, we introduce ConflictQA, a novel benchmark that systematically instantiates conflicts between textual evidence and KG evidence. |
Tianzhe Zhao; Jiaoyan Chen; Shuxiu Zhang; Haiping Zhu; Qika Lin; Jun Liu; |
| 158 | M²GR: Generative User Interest Modeling Via Multi-Granularity Multi-Objective CoT for Industrial Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Recent chain-of-thought (CoT)-based GR methods attempt to address this, but either suffer from information loss during semantic space transformation in explicit reasoning or yield uncontrollable, homogeneous reasoning chains in implicit reasoning.To this end, we propose M2GR, a Generative user interest modeling method using a Multi-granularity Multi-objective CoT for industrial RSs. |
Yuqi Zhang; Jingwen Shi; Wen Shi; Jia Li; Zhen Chen; Jing Zhou; Dongyue Wang; Xiwei Zhao; Sulong Xu; |
| 159 | Diagnosing LLM Reranker Behavior Under Fixed Evidence Pools Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce a controlled diagnostic for reranking that uses Multi-News clusters as fixed evidence pools. |
Baris Arat; Emre Sefer; |
| 160 | Generative Auto-Bidding with Unified Modeling and Exploration Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This limitation leads to inefficient exploration and significantly increases the financial risk for advertising platforms. To bridge this gap, we propose a new framework named Generative Auto-Bidding with Uni fied Mod eling and Exploration (baby), which synergistically integrates directed exploration with a safe fallback mechanism. |
Mingming Zhang; Feiqing Zhuang; Na Li; Shengjie Sun; Xiaowei Chen; Junxiong Zhu; Fei Xiao; Keping Yang; Lixin Zou; Chenliang Li; |
| 161 | FedMosaic: Federated Retrieval-Augmented Generation Via Parametric Adapters Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present FedMosaic, the first federated RAG framework built on parametric adapters. |
Zhilin Liang; Yuxiang Wang; Zimu Zhou; Hainan Zhang; Boyi Liu; Yongxin Tong; |
| 162 | EvoMem: Timeline-Grounded Validity Adjudication for Non-Monotonic Multi-Agent Memory Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present EvoMem, a lightweight plug-in for post-retrieval evidence structuring that converts retrieved interaction traces into an evolutionary timeline: it pools backend evidence, selects salient timepoints under a fixed budget, and summarizes each timepoint to trace updates and suppress outdated answers. |
Zhenhua Wang; Chunlei Wang; Hanchen Luo; Yike Gao; Bang Wang; |
| 163 | Bridging Behavior and Semantics for Time-aware Cross-Domain Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address these, we propose a novel framework that bridges Behavior and Semantics for Time-aware Cross-Domain Sequential Recommendation (BST-CDSR ). |
Zhida Qin; Zemu Liu; Haoyan Fu; Chong Zhang; Tianyu Huang; Yidong Li; Gangyi Ding; |
| 164 | Optimizing Marketing Subsidies Via Counterfactual Learning with Asymmetric Reward Function Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper argues that optimal allocation fundamentally depends on predicting the expected optimal subsidy, a challenge distinct from conventional treatment effect estimation or causal decision-making, which existing approaches fail to address. To fill this gap, we introduce a two-stage Counterfactual optimal subsidy Learning method with an Asymmetric reward (CoLA). |
Xiang Li; Yanghao Xiao; Chunyuan Zheng; Qian Zou; Cheng Bing; Wei Lin; Haoxuan Li; Zhouchen Lin; |
| 165 | User Activity Modeling Under Inflated Distribution Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing methods either rely on the assumption of a normal distribution, neglecting the optimization objectives to focus purely on designing complex model architectures for user feature extraction, which renders them ineffective in non-normally distributed real-world industrial scenarios, or they have explored modeling non-normal distributions, but still struggle to directly operate on the discrete data space in user activity modeling and mitigate the two-sided inflation prevalent in industrial data. Based on this finding, we propose a simple yet effective user activity modeling method under two-sided inflated distribution. |
Xiang Li; Zhao-Yu Zhang; Chunyuan Zheng; Qingying Chen; Huiyou Jiang; Haoxuan Li; Zhouchen Lin; |
| 166 | RecNextEval: A Reference Implementation for Temporal Next-Batch Recommendation Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, recent critical examinations of RecSys evaluation protocols have raised concerns regarding the validity of existing evaluation pipelines. In this demonstration, we present RecNextEval, a reference implementation of an evaluation framework specifically designed for next-batch recommendation. |
Tze-Kean Ng; Joshua Teng-Khing Khoo; Aixin Sun; |
| 167 | Agentic Spatio-Temporal Grounding Via Collaborative Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Weakly-supervised variants mitigate annotation costs but remain constrained by the dataset-level train-and-fit paradigm with an inferior performance. To address these challenges, we propose the Agentic Spatio-Temporal Grounder (ASTG) framework for the task of STVG in an open-world and zero-shot setting. |
Heng Zhao; Yew-Soon Ong; Joey Tianyi Zhou; |
| 168 | Code-Based English Models Reveal Surprising Performance on Chinese QA Pair Extraction Task Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper explores advancements in automated Question-Answer (QA) extraction using large language models (LLMs), addressing challenges in transforming unstructured text into high-quality, retrievable QA pairs. |
Jiajun Yu; Linghan Zheng; Hui Liu; Jiayuan Dong; Yue Shen; Zhiwei Liu; Yaozhen Liang; Yong Li; Haishuai Wang; |
| 169 | SPRINT: Scalable and Predictive Intent Refinement for LLM-Enhanced Session-based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose SPRINT, a scalable SBR framework that incorporates reliable and informative intents while ensuring high efficiency in both training and inference. |
Gyuseok Lee; Wonbin Kweon; Zhenrui Yue; Yaokun Liu; Yifan Liu; Susik Yoon; Dong Wang; SeongKu Kang; |
| 170 | CroSearch-R1: Better Leveraging Cross-lingual Knowledge for Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To better leverage multilingual knowledge, we propose CroSearch-R1, a search-augmented reinforcement learning framework to integrate multilingual knowledge into the Group Relative Policy Optimization (GRPO) process. |
Rui Qi; Fengran Mo; Sijin Lu; Yufeng Chen; Jian-Yun Nie; Kaiyu Huang; |
| 171 | Do Simulated Users Need to Remember? Analyzing The Impact of Memory Models in Conversational Search Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we ask: Do simulated users need to remember? |
Nailia Mirzakhmedova; Marcel Gohsen; Johannes Kiesel; Matthias Hagen; Benno Stein; |
| 172 | Factorized Latent Reasoning for LLM-based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Factorized Latent Reasoning (FLR), a novel framework for LLM-based sequential recommendation that decomposes latent reasoning into multiple disentangled preference factors. |
Tianqi Gao; Chengkai Huang; Zihan Wang; Cao Liu; Ke Zeng; Lina Yao; |
| 173 | One Adapts to Any: Meta Reward Modeling for Personalized LLM Alignment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To ensure robustness, we introduce the Robust Personalization Objective (RPO), which places greater emphasis on hard-to-learn users during meta optimization. |
Hongru Cai; Yongqi Li; Tiezheng Yu; Fengbin Zhu; Wenjie Wang; Fuli Feng; Wenjie Li; |
| 174 | Content-Based Dataset Knowledge Graphs for Dataset Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce CoDaKG, the first content-enriched dataset knowledge graph aligned with the NTCIR benchmark. |
Qing Shi; Jing He; Xintian Pan; Jialiang Wan; Gong Cheng; |
| 175 | Sparton: Fast and Memory-Efficient Triton Kernel for Learned Sparse Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose SPARTON, a fast—memory-efficient—Triton kernel tailored for the LM head in LSR models. |
Thong Nguyen; Cosimo Rulli; Franco Maria Nardini; Rossano Venturini; Andrew Yates; |
| 176 | Towards Inclusive Retrieval-Augmented Generation: Challenges and Opportunities for Cognitively Impaired Users Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We advocate for inclusive IR through our cognitively-adaptive RAG that supports cognitive independence in dementia care, and call on the IR community to advance accessibility and equity through inclusive retrieval system design. |
Claire Rogers; Asmaa Z. A. M. Alqadri; Fiona A. Beaton; Yashar Moshfeghi; |
| 177 | Beyond Static Diffusion: Explicitly Modeling Temporal Patterns in Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Recent diffusion-based generative recommenders show promise in capturing complex dependencies, but they typically treat temporal context as an external conditioning signal rather than integrating temporal transitions into the diffusion dynamics. In this paper, we introduce TDRec (Temporally-aware Diffusion for sequential Recommendation), a novel framework that integrates temporal progression into both forward and reverse processes: at each diffusion step, a position’s latent is updated by noise injection and by mixing with its preceding latent. |
Yao Wu; Chengyi Liu; Wenqi Fan; Rui Zhang; |
| 178 | Context Convergence Improves Answering Inferential Questions Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While Large Language Models (LLMs) are widely used in open-domain Question Answering (QA), their ability to handle inferential questions—where answers must be derived rather than directly retrieved—remains still underexplored. This study investigates how the structure and quality of passages influence LLM performance on such questions. |
Jamshid Mozafari; Bhawna Piryani; Adam Jatowt; |
| 179 | Pretraining Exposure Explains Popularity Judgments in Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We provide the first direct, large-scale analysis of popularity bias grounded in fully observable pretraining data. |
Jamshid Mozafari; Bhawna Piryani; Adam Jatowt; |
| 180 | Beyond Static Best-of-N: Bayesian List-wise Alignment for LLM-based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address this, BoN Alignment aims to distill the search capability into the model itself, yet current approaches suffer from two critical limitations: (1) Indiscriminate Supervision, where the static reference fails to distinguish the relative quality of candidates exceeding its empirical range, leading to a loss of ranking guidance; and (2) Gradient Decay, where the effective supervision signal rapidly diminishes as the evolving policy improves, resulting in inefficient optimization. To overcome these challenges, we propose BLADE (Bayesian List-wise Alignment via Dynamic Estimation). |
Ruijun Chen; Chongming Gao; Jiawei Chen; Weiqin Yang; Xiangnan He; |
| 181 | Hypencoder Revisited: Reproducibility and Analysis of Non-Linear Scoring for First-Stage Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we conduct a reproducibility study of the Hypencoder and extend the original analysis in three directions. |
Arne Eichholtz; Yongkang Li; Jutte Vijverberg; Tobias Groot; Mohammad Aliannejadi; |
| 182 | The Powerless Noise: How Experimental Settings Shape The Reported Power of Noise Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we reproduce the main findings of Cuconasu et al. and evaluate the robustness of the effect under extended experimental settings. |
Micha\l{} Mazuryk; Fleur Dolmans; Louis Gehringer; Ina Klaric; Jia-Huei Ju; Mohammad Aliannejadi; |
| 183 | CalmSet: A Domain-Specific Test Collection for Affective Music Retrieval for Children with ASD Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce CalmSet, a test collection for emotion-tagged music retrieval and recommendation in a therapeutic context for children with Autism Spectrum Disorder (ASD). |
Abhishek Karwankar; Liam Stapley; Daniel Stevens; Matthew Louis Mauriello; |
| 184 | KuaiLive: A Real-time Interactive Dataset for Live Streaming Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, research progress in academia has been hindered by the lack of publicly available datasets that accurately reflect the dynamic nature of live streaming environments. To address this gap, we introduce KuaiLive, the first real-time, interactive dataset collected from Kuaishou, a leading live streaming platform in China with over 400 million daily active users. |
Changle Qu; Sunhao Dai; Ke Guo; Xiao Zhang; Liqin Zhao; Shijun Wang; Yanan Niu; Lantao Hu; Han Li; Jun Xu; |
| 185 | Pretrain-then-Adapt: Uncertainty-Aware Test-Time Adaptation for Text-based Person Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose a new Pretrain-then-Adapt paradigm that eliminates reliance on extensive target-domain supervision through an offline test-time adaptation manner, enabling dynamic model adaptation using only unlabeled test data with minimal post-train time cost. |
Jiahao Zhang; Shaofei Huang; Yaxiong Wang; Zhedong Zheng; |
| 186 | MKG-RAG: Leveraging Multimodal Knowledge Graphs in Retrieval-Augmented Generation for Knowledge-intensive VQA Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, this paper proposes mKG-RAG, a novel retrieval-augmented generation framework built upon multimodal KGs for knowledge-intensive VQA tasks. |
Xu Yuan; Liangbo Ning; Qingqing Ye; Wenqi Fan; Qing Li; |
| 187 | Distribution-aware Re-representations for Multi-Scenario Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose a Distribution-aware Re-representations (DAR) method for MSR. |
Qi Sun; Yulin Xu; Zelin Wang; Xiaoyu Kang; Keyan Jin; Jiechao Gao; |
| 188 | Inferring Targets from Calibrated Hesitations Via Mutual Information Maximization in Multi-Behavior Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Consequently, models cannot distinguish true disinterest from intended but hesitant behavior, which introduces substantial noise into preference modeling. To address this issue, we propose a novel framework named Calibrated Hesitation Analysis for Multi-Behavior Recommendation via Mutual Information Maximization (CHARM). |
Cheng Li; Yong Xu; Suhua Tang; Xin He; Jianfeng Sun; Jinde Cao; |
| 189 | Retrieval-Augmented Multimodal Model for Fake News Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Nevertheless, this direction presents two significant challenges: (1) Failure to Capture Cross-Instance Narrative Consistency: existing models usually evaluate each news in isolation, fail to capture cross-instance narrative consistency, and thus struggle to address the spread of cluster-based fake news driven by social media; (2) Lack of Domain-Specific Knowledge for Reasoning: conventional models, which rely solely on knowledge encoded in their parameters during training, struggle to generalize to new or data-scarce domains (e.g., emerging events or niche topics). To tackle these challenges, we introduce Retrieval-Augmented Multimodal Model for Fake News Detection (RAMM). |
Yiheng Li; Weihai Lu; Hanyi Yu; Yue Wang; |
| 190 | ProMax: Exploring The Potential of LLM-derived Profiles with Distribution Shaping for Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Moreover, existing studies predominantly rely on nonlinear alignment and fusion strategies to incorporate these profiles, which often lead to semantic loss and fail to fully exploit their potential. To address these limitations, we revisit profiles from a retrieval perspective and propose a simple yet effective recommendation framework built upon distribution shaping (ProMax) in this paper. |
Yi Zhang; Yiwen Zhang; Kai Zheng; Tong Chen; Hongzhi Yin; |
| 191 | RegionSLM: Region-aware Question Answering on Document Screenshots Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, most Screenshot Language Models (SLMs) encode the entire page holistically and rely on implicit attention to ”find” relevant content, which limits both accuracy and efficiency. We present RegionSLM, a region-aware SLM designed to explicitly connect the question to its supporting regions. |
Chao Wang; Hehe Fan; Huichen Yang; Sarvnaz Karimi; Lina Yao; Yi Yang; |
| 192 | SSR: Structured Subgraph Retrieval for Temporal Knowledge Graph Question Answering with LLMs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Moreover, selecting top-n facts based on semantic similarity inevitably introduces a large amount of irrelevant noise into the LLM input, which degrades reasoning performance. To address these limitations, we propose SSR, a Structured Subgraph Retrieval framework for TKGQA with LLMs. |
Ying Zhang; Li Zhang; Wenya Guo; Shilong Ping; Xinying Qian; |
| 193 | Discrete Preference Learning for Personalized Multimodal Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To tackle these, we present a two-stage framework called Discrete Preference learning for Personalized Multimodal Generation (DPPMG). |
Yuting Zhang; Ying Sun; Dazhong Shen; Ziwei Xie; Feng Liu; Changwang Zhang; Xiang Liu; Jun Wang; Hui Xiong; |
| 194 | Beyond Chunk-Then-Embed: A Comprehensive Taxonomy and Evaluation of Document Chunking Strategies for Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Recent research has proposed several concurrent approaches, including LLM-guided methods (e.g., DenseX and LumberChunker) and contextualized strategies (e.g., Late Chunking), which generate embeddings before segmentation to preserve contextual information. |
Yongjie Zhou; Shuai Wang; Bevan Koopman; Guido Zuccon; |
| 195 | STAR: Staged Training with Aligned Reinforcement Learning and Multi-Faceted Distillation for Interpretable E-commerce Relevance Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Knowledge distillation offers a promising solution, yet current methods force an undesirable trade-off: sacrificing the very interpretability that makes LLMs powerful, or relying on expensive, unscalable human-annotated rationales. To address this, we propose STAR—Staged Training with Aligned Reinforcement Learning and Multi-Faceted Distillation, a progressive framework that follows a reasoning, ranking, and transfer pipeline to imbue dense models with both high performance and interpretability. |
Chenxu Wang; Jianzhi Shao; Chi Zhang; Tao Zhang; |
| 196 | TRACE: Term-level Reasoning And Chain-of-thought Enhanced Distillation for E-commerce Multi-modal Relevance Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While Large Language Models (LLMs) excel at such reasoning, their high computational overhead makes direct online deployment infeasible. To bridge this gap, we propose TRACE (Term-level Reasoning And Chain-of-thought Enhanced distillation), a framework designed for deploying advanced reasoning capabilities at scale. |
Chenxu Wang; Chi Zhang; Fangyi Liang; Jianzhi Shao; Manyi Wang; Tao Zhang; |
| 197 | RCLRec: Reverse Curriculum Learning for Modeling Sparse Conversions in Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While recent behavior-aware GR models encode behavior types and employ behavior-aware attention to highlight decision-related intermediate behaviors, they still rely on standard attention over the full history and provide no additional supervision for conversions, leaving conversion sparsity largely unresolved. To address these challenges, we propose RCLRec, a reverse curriculum learning–based GR framework for sparse conversion supervision. |
Yulei Huang; Hao Deng; Haibo Xing; Jinxin Hu; Chuanfei Xu; Zulong Chen; Yu Zhang; Xiaoyi Zeng; |
| 198 | Large-Scale Online Learning for Generative List Recommendation in E-commerce: An Environment Policy Optimization Approach Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose Environment Policy Optimization (EPO), a novel GLR model that fundamentally reshapes policy learning by exploiting the differentiability of the environment within the Generator-Evaluator framework. |
Yuan Wang; Zhiyu Li; Ang Gao; Changshuo Zhang; Xiao Zhang; Jun Xu; Quan Lin; |
| 199 | HALO: Hyperbolic Adaptation Via Lift Overlay for Hierarchy-Aware Cross-Modal Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose HALO (Hyperbolic Adaptation via Lifted Overlay), a lightweight hyperbolic retrieval framework that equips Euclidean-pretrained VLMs with hierarchy-aware search at low computational overhead, without compromising retrieval performance. |
Teng Long; Andrew Yates; |
| 200 | Denoise and Align: Diffusion-Driven Foreground Knowledge Prompting for Open-Vocabulary Temporal Action Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing methods struggle to mitigate the semantic imbalance between concise, abstract action labels and rich, complex video contents, inevitably introducing semantic noise and misleading cross-modal alignment. To address this challenge, we propose DFAlign, the first framework that leverages diffusion-based denoising to generate foreground knowledge for the guidance of action–video alignment. |
Sa Zhu; Wanqian Zhang; Lin Wang; Jinchao Zhang; Cong Wang; Bo Li; |
| 201 | Fast and Faithful: Real-Time Verification for Long-Document Retrieval-Augmented Generation Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present the design of a real-time verification component integrated into a production RAG pipeline that enables full-document grounding under latency constraints. |
Xunzhuo Liu; Bowei He; Xue Liu; Haichen Zhang; Huamin Chen; |
| 202 | A Standardized Re-evaluation of Conversational Recommender Systems on The ReDial Dataset Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: These comparisons are further complicated by confounding factors such as the choice of the underlying large language model (LLM) and the use of external data sources. In this work, we revisit seven prominent CRS methods across three architectural families and evaluate them under standardized conditions. |
Ivica Kostric; Krisztian Balog; |
| 203 | EviRAG: Evidence-Guided Retrieval-Augmented Generation for Medical Vision-Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose an evidence-guided retrieval-augmented framework EviRAG that decomposes retrieval into structured and unstructured alignment levels. |
Yiyang Gu; Jiayue Fan; Kaili Liu; Bohan Wu; Binqi Chen; Zequn Liu; Zhiping Xiao; Rong-Cheng Tu; Xiao Luo; Ming Zhang; |
| 204 | SynDiSC: High-Quality Tabular Data Synthesis with Distributional and Semantic Consistency Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, this task remains challenging due to two key factors: (1) distribution complexity : imbalanced and skewed data make it challenging to learn the data distribution accurately; and (2) semantic coherence : implicit relationships and logical dependencies among fields must be preserved to ensure valid and meaningful synthetic samples. To address these issues, we propose SynDiSC, a high-quality tabular data synthesis approach that enforces both distributional and semantic consistency. |
Fan Wu; Haoye Pan; Hao Wu; Kai Qian; Shucheng Li; Feng Lyu; |
| 205 | A Parametric Memory Head for Continual Generative Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We show that sequential adaptation improves retrieval on newly added documents but substantially degrades performance on earlier slices, exposing a pronounced stability–plasticity trade-off. To address this, we propose post-adaptation memory tuning (PAMT), a memory-only stabilization stage that augments an adapted model with a modular parametric memory head (PMH). |
Kidist Amde Mekonnen; Yubao Tang; Maarten de Rijke; |
| 206 | Lost in Decoding? Reproducing and Stress-Testing The Look-Ahead Prior in Generative Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Using the authors’ released checkpoint and identifier/trie artifacts under the reported decoding setup, we reproduce the main effectiveness results on MS~MARCO Dev and TREC-DL 2019/2020, and corroborate the reported beam-size–latency trade-off in our hardware setting. Beyond reproduction, we introduce plan drift diagnostics that quantify how intent-preserving query variations, including misspellings, reordering, synonym substitutions, paraphrases, and naturality shifts, alter the planner’s top-n candidate set and highest-weight planner tokens, and how these changes affect guided decoding. |
Kidist Amde Mekonnen; Yongkang Li; Yubao Tang; Simon Lupart; Maarten de Rijke; |
| 207 | Struct-Align: Zero-Shot Text-to-3D Scene Retrieval Via Locality-Aware Structural Alignment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose Struct-Align, a foundation-model-driven framework for zero-shot T3SR that eliminates the need for paired training data. |
Xiong Li; Yikang Yan; Zhenyu Wen; Jie Su; Qi Chen; Zhen Hong; |
| 208 | CRED: Calibrated Relational Enhanced Distillation for LLM-Based Pointwise Reranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To develop lightweight yet high-performance LLM-based pointwise rerankers through knowledge distillation, we identify two critical limitations: teachers often yield over-smoothed and inaccurate supervision on hard negative samples, thereby hindering the student’s optimization; furthermore, traditional methods underutilize the relevance score differences between candidates, which are crucial for ranking tasks. To address these challenges, we propose CRED (Calibrated Relational Enhanced Distillation), which integrates Adaptive Teacher Calibration (ATC) to calibrate teacher predictions and amplify score margins, while employing Preference Relation Alignment (PRA) to align the distributional patterns of relevance score differences, enabling the student to capture precise ranking structures. |
Qingran Yang; Wenxuan Zhang; Yuting Wang; Xue Li; Lanshun Nie; |
| 209 | ADaFuSE: Adaptive Diffusion-generated Image and Text Fusion for Interactive Text-to-Image Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing frameworks fuse multi-modal user feedback by simple embedding addition. In this work, we show that this basic fusion strategy indiscriminately incorporates generative noise produced by the diffusion model, leading to performance degradation for up to 55.62\% of samples. |
Zhuocheng Zhang; Xingwu Zhang; Kangheng Liang; Guanxuan Li; Richard McCreadie; Zijun Long; |
| 210 | GATHER: Convergence-Centric Hyper-Entity Retrieval for Zero-Shot Cell-Type Annotation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose GATHER (Graph-Aware Traversal with Hyper-Entity Retrieval), a convergence-centric retriever tailored to hyper-entity queries. |
Zhonghui Zhang; Feng Jiang; Shaowei Qin; Jiahao Zhao; Min Yang; |
| 211 | SpecTran: Spectral-Aware Transformer-based Adapter for LLM-Enhanced Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Recent advances in Large Language Models (LLMs) have inspired a surge of research that encodes item textual information with high-dimensional semantic embeddings, and designs transformation methods to inject such embeddings into SR models. These embedding transformation strategies can be categorized into two types, both of which exhibits notable drawbacks: 1) adapter-based methods suffer from pronounced dimension collapse, concentrating information into a few dominant dimensions; 2) SVD-based methods are rigid and manual, considering only a few principal spectral components while discarding rich information in the remaining spectrum.To address these limitations, we propose SpecTran, a spectral-aware transformer-based adapter that operates in the spectral domain, attending to the full spectrum to select and aggregates informative components. |
Yu Cui; Feng Liu; Zhaoxiang Wang; Changwang Zhang; Jun Wang; Can Wang; Jiawei Chen; |
| 212 | The Wikidata Query Logs Dataset Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present the Wikidata Query Logs (WDQL) dataset, a dataset consisting of 335k question-query pairs over the Wikidata knowledge graph. |
Sebastian Walter; Hannah Bast; |
| 213 | DLLM-Searcher: Adapting Diffusion Language Model for Efficient Search Agents Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose DLLM-Searcher, an optimization framework for dLLM-based Search Agents. |
Jiahao Zhao; Shaoxuan Xu; Zhongxiang Sun; Fengqi Zhu; Jingyang Ou; Yuling Shi; Chongxuan Li; Xiao Zhang; Jun Xu; |
| 214 | FedMM: Federated Collaborative Signal Quantization for Multi-Market CTR Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Additionally, because ID spaces are disjoint across markets, embedding-based aggregation strategies become ineffective. To overcome these challenges, we propose a federated collaborative signal quantization (FedMM) method for multi-market click-through rate (CTR) prediction. |
Jun Zhang; Dugang Liu; Xing Tang; Xiuqiang He; Zhong Ming; |
| 215 | Learning Decomposed Contextual Token Representations from Pretrained and Collaborative Signals for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This objective misalignment leads to two key limitations: (i) suboptimal static tokenization, where fixed token assignments fail to reflect diverse usage contexts; and (ii) discarded pretrained semantics, where pretrained knowledge—typically from language model embeddings—is overwritten during recommender training on user interactions. To address these limitations, we propose to learn DEcomposed COntextual Token Representations (DECOR), a unified framework that preserves pretrained semantics while enhancing the adaptability of token embeddings. |
Yifan Liu; Yaokun Liu; Zelin Li; Zhenrui Yue; Ruichen Yao; Yang Zhang; Dong Wang; |
| 216 | Dynamic Memory Forest: Constructing and Tracing Conversational Trajectories for Long-Term Conversation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Summarizing historical conversations has achieved remarkable performance, which, however, loses conversational trajectory and association, making it difficult to precisely combine memories from different sessions in response to current queries. To address this, we propose the Dynamic Memory Forest (DMF), a novel Consolidation-then-Growth framework for long-term open-domain conversation, which simulates the consolidation and growth processes of human memory by dynamically organizing long-term conversation histories into a memory forest of memory trees. |
Cai Ke; Bin Liang; Xin Liu; Yue Yu; Hui Wang; Ruifeng Xu; |
| 217 | Additive Control Variates Dominate Self-Normalisation in Off-Policy Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper provides a definitive answer: we prove that ? |
Olivier Jeunen; Shashank Gupta; |
| 218 | Understanding and Modeling Heterogeneous Search Behavior Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We investigate between-user drivers of query variability through a controlled between-subject, full-factorial user study that manipulates age, gender, and language proficiency across six backstory-driven search tasks. |
Nuha Abu Onq; Chenglong Ma; Mark Sanderson; Falk Scholer; |
| 219 | Revisiting Text Ranking in Deep Research Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Despite search’s essential role in deep research, black-box web search APIs leave the behaviour of established text ranking methods in deep research largely unclear. To fill this gap, we reproduce key findings and best practices for text ranking methods in deep research. |
Chuan Meng; Litu Ou; Sean MacAvaney; Jeff Dalton; |
| 220 | Orcheo: A Modular Full-Stack Platform for Conversational Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: CS researchers currently face two barriers: the lack of a unified framework for efficiently sharing contributions with the community, and the difficulty of deploying end-to-end prototypes needed for user evaluation. We introduce Orcheo, an open-source platform designed to bridge this gap. |
Shaojie Jiang; Svitlana Vakulenko; Maarten de Rijke; |
| 221 | Post-hoc Provider Fairness Adaptation Via Hierarchical Exposure Alignment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Yet, most existing methods either incorporate fairness constraints during model training, requiring expensive retraining when fairness objectives change, or rely on post-hoc reranking with fixed criteria, which lacks adaptability to diverse fairness requirements. To overcome these limitations, we propose Post-hoc Fairness Adaptation (PFA), a lightweight framework that equips a frozen recommender with a fairness adapter, enabling flexible fairness control without retraining the backbone model. |
Jingzhi Li; Zhiyong Cheng; Richang Hong; Meng Wang; |
| 222 | One-for-All Community Search on Unseen Graphs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While recent learning-based methods improve retrieval effectiveness via graph representation learning, they follow a ”one-use-one-train” paradigm that requires retraining or fine-tuning for each target graph, leading to high data dependency, high training costs, and limited generalization. To handle this, we propose OFA-CS, a ”one-for-all” community search framework trained once on source datasets and directly deployed to arbitrary unseen graphs without retraining or fine-tuning, while preserving strong performance. |
Mo Li; Zhaosong Zhao; Linlin Ding; Renata Borovica-Gajic; Zhongming Yao; Jianxin Li; |
| 223 | Negotiating The Punchline: Contextual Meme Understanding Via Discrete Semantic Energy Minimization Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Lacking mechanisms to measure and correct misalignment, these static models inevitably allow initial perceptual failures to cascade into irreversible hallucinations. To address this, we propose Semantic Energy Entropy Descent (SEED), a framework that reformulates contextual meme understanding as an energy minimization problem within a discrete semantic space. |
Bingbing Wang; Zihan Wang; Zhengda Jin; Jing Li; Ruifeng Xu; Min Zhang; |
| 224 | From Single- to Cross-Document: Benchmarking Multi-Granularity Event Analysis of Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While large language models (LLMs) have preliminarily achieved promising performance in part of these tasks individually, their capability in event analysis still lacks comprehensive understanding due to restricted document granularity, task designs, and data source of existing benchmarks. To address these limitations, we introduce MiGUE-Bench, a systematic benchmark for assessing the performance of LLMs in multi-granularity event analysis. |
Tao Wen; Shuai Shao; Pei Ke; Xu Han; Jie Zou; Guannan Li; Tao Tian; Jinjie Qiu; Lan Wang; Ke Qin; |
| 225 | Auto-ARGUE: LLM-Based Report Generation Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Accordingly, we introduce Auto-ARGUE, a robust LLM-based implementation of the recently proposed ARGUE framework for report generation evaluation. |
William Walden; Marc Mason; Orion Weller; Laura Dietz; John M. Conroy; Neil Molino; Hannah Recknor; Bryan Li; Gabrielle Liu; Yu Hou; Dawn Lawrie; James Mayfield; Eugene Yang; |
| 226 | Effective Offline LLM and DNN Based Matching, Filtering and Ranking for Search Ads Retrieval in E-Commerce Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: For search Ads retrieval in e-Commerce, when a customer inputs a query, from tens of millions of Ads products, the system needs to quickly retrieve a limited number of candidates that can not only meet the customer’s search intension but also have high conversion probability. This constructs challenges for both retrieval efficiency and retrieval quality on the perspective of relevance and engagement. |
Shuping Ji; Jianguo Yao; Dong Zhang; |
| 227 | Unifying On- and Off-Policy Variance Reduction Methods Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We prove that the standard online Difference-in-Means estimator is mathematically identical to an off-policy Inverse Propensity Scoring estimator equipped with an optimal (variance-minimising) additive control variate. |
Olivier Jeunen; |
| 228 | Cross-Stage Signal Propagation for Negative Filtering in Multi-Stage Ad Delivery Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Cross-Stage Signal Propagation, a unified framework for negative filtering in multi-stage systems. |
Min Zhang; Ganlin Song; Dong Liang; Lizhang Qin; Chao Ma; |
| 229 | ItemRAG: Item-Based Retrieval-Augmented Generation for LLM-Based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Most existing RAG approaches retrieve purchase histories of users similar to the target user; however, these histories often contain noisy or weakly relevant information and provide little or no useful information for candidate items. To address these limitations, we propose ItemRAG, a novel RAG approach that shifts focus from coarse user-history retrieval to fine-grained item-level retrieval. |
Sunwoo Kim; Geon Lee; Kyungho Kim; Jaemin Yoo; Kijung Shin; |
| 230 | Multi-Sourced, Multi-Agent Evidence Retrieval for Fact-Checking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: These limitations lead to overlooking subtle factual correlations between the evidence and the claims to be fact-checked during evidence retrieval, thus causing inaccurate veracity predictions. To address these issues, we propose a Web-enhanced Knowledge Graph retrieval Fact-Checking agentic framework (WKGFC), which exploits authorized open knowledge graph as a core resource of evidence. |
Shuzhi Gong; Richard Sinnott; Jianzhong Qi; Cecile Paris; Preslav Nakov; Zhuohan Xie; |
| 231 | Filling The Gaps: Selective Knowledge Augmentation for LLM Recommenders Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we propose KnowSACKP(Knowledge-aware Selective Augmentation with Comparative Knowledge Probing) to mitigate the knowledge gap problem. |
Jaehyun Lee; Sanghwan Jang; SeongKu Kang; Hwanjo Yu; |
| 232 | A Reproducibility Study of LLM-Based Query Reformulation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a systematic reproducibility and comparative study of ten representative LLM-based query reformulation methods under a unified and strictly controlled experimental framework. |
Amin Bigdeli; Radin Hamidi Rad; Hai Son Le; Mert Incesu; Negar Arabzadeh; Charles L. A. Clarke; Ebrahim Bagheri; |
| 233 | Aspect-Aware Content-Based Recommendations for Mathematical Research Papers Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Recognizing that LLM embeddings of mathematical content alone yield suboptimal representation, we propose AchGNN, an aspect -conditioned heterogeneous GNN that jointly models textual semantics, citation structure, and author lineage. |
Ankit Satpute; Andr\'{e} Greiner-Petter; Noah Gie\ss{}ing; Olaf Teschke; Moritz Schubotz; Akiko Aizawa; Bela Gipp; |
| 234 | Latent Retrieval Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: More fundamentally, both paradigms rely on explicit discrete representations tokens or parameters that may introduce information bottlenecks and hinder seamless knowledge integration. To address these challenges, we introduce Latent RAG, a novel paradigm that performs knowledge injection entirely within the continuous latent space. |
Shu Zhou; Junan Chen; Rui Ling; Tao Fan; Hao Wang; |
| 235 | Calibrating Uncertainty with Cross-Model Consistency for LLM Hallucination Mitigation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we observe that answers agreed upon by multiple models are significantly more likely to be correct-a manifestation of the wisdom of crowds principle. |
Shu Zhou; Rui Ling; Junan Chen; Tao Fan; Hao Wang; |
| 236 | SCORE-RAG: Self-Correcting Exploration-Exploitation Retrieval for Multi-hop Question Answering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This assumption proves inadequate for multi-hop questions, where comprehending the query itself often requires retrieval support, creating a chicken-and-egg dilemma between query understanding and information retrieval. To address this challenge, we propose SCORE-RAG Self-COrrecting Exploration-Exploitation REtrieval, a novel framework inspired by the explore-exploit paradigm in decision theory. |
Shu Zhou; Rui Ling; Junan Chen; Tao Fan; Hao Wang; |
| 237 | Why Knowledge Distillation Fails to Scale in Neural Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we hypothesize that the teacher model’s capacity acts as an information bottleneck, limiting how much large student models can learn. |
Shu Zhou; Rui Ling; Junan Chen; Tao Fan; Hao Wang; |
| 238 | CS3: Efficient Online Capability Synergy for Two-Tower Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Capability Synergy (CS3), an efficient online framework that strengthens two-tower retrievers while preserving real-time constraints. |
Lixiang Wang; Shaoyun Shi; Peng Wang; Wenjin Wu; Peng Jiang; |
| 239 | Prompt-Unknown Promotion Attacks Against LLM-based Sequential Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we investigate the item promotion attack in LLM-SRSs under a more realistic setting where both the system prompt and victim model are unknown to the attacker, and propose a Prompt-Unknown Dual-poisoning Attack (PUDA) framework. |
Yuchuan Zhao; Tong Chen; Junliang Yu; Zongwei Wang; Lizhen Cui; Hongzhi Yin; |
| 240 | Multimodal Large Language Models with Adaptive Preference Optimization for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Yet, two core challenges remain: 1) Imbalanced sample hardness, where random negative sampling causes overfitting on easy examples and under-training on hard ones; 2) Cross-modal semantic bias, where the fixed reference model in DPO prevents the policy model from correcting modality misalignments—especially over long sequences. To address these issues, we propose a Multimodal LLM framework that integrates Hardness-aware and Noise-regularized preference optimization for Recommendation (HaNoRec). |
Yu Wang; Yonghui Yang; Le Wu; Yi Zhang; Fei Liu; Richang Hong; |
| 241 | Booking Funnel and Substitution-Aware User Behavior Modeling for Demand Prediction and Joint Room Pricing Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose BFSNet, a Booking Funnel and Substitution-aware neural network that decomposes booking demand into hotel click and booking conversion, enabling interpretable and stage-wise learning of user behavior. |
Zhikang Fan; Zihao Chen; Pin Gao; Ruohan Zhan; Shaowen Zhang; Xingrui Li; Jianpei Wen; Su Zhao; Shuai Chen; Wei Lin; |
| 242 | Policy-Grounded Dynamic Facet Suggestions for Job Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose a policy-grounded, retrieval-augmented ranking framework for facet suggestion, comprising offline taxonomy curation, embedding-based retrieval of top-K candidates, and a distilled small language model (SLM) based candidate scoring. |
Dan Xu; Baofen Zheng; Qianqi Shen; Jianqiang Shen; Wenqiong Liu; Chunnan Yao; Ping Liu; Rajat Arora; Kevin Kao; Hsiang Lin; Wanjun Jiang; Yusuke Takebuchi; Jingwei Wu; Wenjing Zhang; |
| 243 | Unified Semantic Modeling Framework for Large-Scale Job Understanding at LinkedIn Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present a unified semantic modeling framework powered by a small language model (SLM) to address the challenges. |
Dan Xu; Baofen Zheng; Jianqiang Shen; Qi Xiao; Benjamin Hoan Le; Wen Pu; Saurabh Gupta; Ran Zhou; Neha Saraf; Alice Leung; Qianqi Shen; Liangjie Hong; Jingwei Wu; Wenjing Zhang; |
| 244 | Sparse Contrastive Learning for Content-Based Cold Item Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, such approaches can be limited by the fundamental information gap between CF signals and content features. In this work, we propose to avoid this limitation with purely content-based modeling of cold items, i.e. without alignment with CF user or item embeddings. |
Gregor Meehan; Johan Pauwels; |
| 245 | Query-Attention Dual-Stream Framework with Cross-Category Transfer for Efficient Fine-Grained Interest Pre-Ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, pre-ranking still faces challenges of behavioral sparsity, limited interest diversity, and computational latency. We propose the Query-Attention Dual-Stream (QADS) framework to address these issues. |
Huimu Wang; Xujun Liu; Yiming Qiu; Zhenlin He; Enqiang Xu; Yihao Wang; Jinyuan Zhao; Guangtao Nie; Songlin Wang; Mingming Li; |
| 246 | Bridging The Gap: Generative Retrieval Via Query-to-Multi-Span Framework for Effective E-commerce Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Furthermore, generating accurate targets from brief queries against noisy, loosely structured item titles remains a practical challenge. To address these issues, we propose a Query-to-Multi-Span generative retrieval framework tailored for E-commerce. |
Huimu Wang; Yiming Qiu; Xingzhi Yao; Guangtao Nie; Zuxu Chen; Zhenlin He; Songlin Wang; Guoyu Tang; Sulong Xu; Jingwei Zhuo; Mingming Li; |
| 247 | Towards Efficient and Generalizable Retrieval: Adaptive Semantic Quantization and Residual Knowledge Transfer Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: On the other hand, data-sparse tail items especially cold-start items are prone to semantic fragmentation during quantization; they are often mapped as isolated discrete points, which severely hinders their ability to generalize. To address this issue, we propose the Anchored Curriculum with Sequential Adaptive Quantization (SA2CRQ) framework. |
Huimu Wang; Xingzhi Yao; Yiming Qiu; Qinghong Zhang; Haotian Wang; Yufan Cui; Songlin Wang; Sulong Xu; Mingming Li; |
| 248 | MedNQS: Medical State-Aware Dual-Stage Next-Turn Question Suggestion for Online Medical Consultations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present MedNQS, a medical state-aware dual-stage NQS system for online medical consultations. |
Yue He; Dongsheng Bi; Minghui Yang; Jian Wang; Jinjie Gu; Junwei Liu; |
| 249 | IAgentBench: Benchmarking Sensemaking Capabilities of Information-Seeking Agents on High-Traffic Topics Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Yet many widely used QA benchmarks remain answerable by retrieving a single relevant passage, making them poorly suited for measuring cross-source sensemaking, such as integrating evidence, tracking causal links, and resolving dependencies across facets of a topic. We present iAgentBench, a dynamic ODQA benchmark that targets these higher-level information needs while keeping questions natural and grounded in realistic information-seeking behavior. |
Preetam Prabhu Srikar Dammu; Arnav Palkhiwala; Tanya Roosta; Chirag Shah; |
| 250 | From Clues to Generation: Language-Guided Conditional Diffusion for Cross-Domain Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: For these users, the absence of explicit alignment signals makes fine-grained preference transfer intrinsically difficult. To address this challenge, this paper proposes Language-Guided Conditional Diffusion for CDR (LGCD), a novel framework that integrates Large Language Models (LLMs) and diffusion models for inter-domain sequential recommendation. |
Ziang Lu; Lei Sang; Lin Mu; Yiwen Zhang; |
| 251 | Counterfactual Multi-task Learning for Delayed Conversion Modeling in E-commerce Sales Pre-Promotion Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Our model incorporates three key innovations: (i) A multi-task architecture that jointly models direct and delayed conversions using historical pre-promotion data; (ii) A personalized user behavior gating module to mitigate data sparsity issues during brief pre-promotion periods; (iii) A counterfactual causal approach to model the transition probability from add-to-cart (ATC) to delayed conversion.Extensive experiments demonstrate that CM-DCM outperforms state-of-the-art delayed CVR models in pre-promotion scenarios. |
Xin Song; Kaiyuan Li; Jinxin Hu; |
| 252 | One-Pass Decoding for Generative Recommendation with WFST-Constrained A* Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose OneGR, a novel generative recommendation framework that formulates SID prediction as a one-pass, structure-constrained decoding problem. |
Puji Wang; Yingchen Zhang; Ruqing Zhang; Jiafeng Guo; Xueqi Cheng; |
| 253 | The Vulnerability of LLM Rankers to Prompt Injection Attacks: You Are to [MARK] This Paper As The Best Paper Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present a comprehensive empirical study of jailbreak prompt attacks against LLM rankers. |
Yu Yin; Shuai Wang; Bevan Koopman; Guido Zuccon; |
| 254 | RQ-GMM: Residual Quantized Gaussian Mixture Model for Multimodal Semantic Discretization in CTR Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose RQ-GMM (Residual Quantized Gaussian Mixture Model), which introduces probabilistic modeling to better capture the statistical structure of multimodal embedding spaces. |
Ziye Tong; Jiahao Liu; Weimin Zhang; Hongji Ruan; Derick Tang; Zhanpeng Zeng; Qinsong Zeng; Peng Zhang; Tun Lu; Ning Gu; |
| 255 | Is It Novel and Why? Fine-Grained Patent Novelty Prediction Based on Passage Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we introduce FiNE-Patents (Fine-grained Novelty Examination of Patents), a novel dataset comprising 3,658 first patent claims annotated with fine-grained, feature-level prior art references extracted from European Search Opinion (ESOP) documents. |
Valentin Knappich; Anna H\{a}tty; Simon Razniewski; Annemarie Friedrich; |
| 256 | TDHGNN: A Temporal Directed Hypergraph Neural Network for Bitcoin Fraud Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To effectively address the unique challenges of Bitcoin fraud detection, such as handling heterogeneous, directional, and time-sensitive transaction patterns within the UTXO model, we propose a novel hypergraph-based representation learning framework. |
Zheng Gong; Shuheng Shen; Changhua Meng; Ying Sun; |
| 257 | Task-Aware Automated User Profile Generation for Recommendation Simulation Using Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Moreover, the scarcity of datasets specifically designed for recommendation simulations has led to heavy reliance on manually crafted profiles, significantly limiting the scalability and generalisability of simulation frameworks across different datasets. To address these challenges, this work proposes an Automated Profile Generation Framework for Recommendation Simulation, APG4RecSim, that constructs realistic, coherent, and robust user profiles with minimal supervision. |
Xinye Wanyan; Chenglong Ma; Danula Hettiachchi; Ziqi Xu; Jeffrey Chan; |
| 258 | FLAME: Condensing Ensemble Diversity Into A Single Network for Efficient Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Frozen and Learnable networks with Aligned Modular Ensemble (FLAME), a novel framework that condenses ensemble-level diversity into a single network for efficient sequential recommendation. |
WooJoo Kim; JunYoung Kim; JaeHyung Lim; SeongJin Choi; SeongKu Kang; HwanJo Yu; |
| 259 | Precision at Scale: An End-to-End Graph-based Framework for Mitigating Network Interference in TikTok A/B Tests Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Conversely, theoretically sound methods remain computationally intractable at production scale. This paper presents a production-ready framework deployed at TikTok, which integrates three core contributions to address these challenges: 1) Learned Interference Graph (LIG): Estimates interference probabilities using dynamic interaction patterns for more context-aware modeling. |
Yuhan Li; Jiayun Ni; Ao Li; Yue Wang; Seng You Paul Chee; Yongmin Hu; Runze Yu; |
| 260 | Failing Forward: Understanding Query Failure in Retrieval, Judgment, and Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Modern information retrieval pipelines combine retrieval, LLM-based generation, and LLM-based judgment, and a poor outcome may originate in any of the three stages. |
Seyed Mohammad Hosseini; Negar Arabzadeh; Mohammad Hossein Saliminabi; Dimitrios Androutsos; Morteza Zihayat; Ebrahim Bagheri; |
| 261 | OPS: An Order-Preserving Sorting Network for Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address the first issue, we propose an Order-Preserving Sorting network (OPS), which combines a relaxation operator with random hard swaps to explicitly control the approximation error between the predicted soft permutation matrix and the exact hard permutation matrix. |
Chao Wang; Yongxiang Tang; Guikai Luan; Kaiyuan Li; Yanhua Cheng; Xialong Liu; Shu Wu; Peng Jiang; |
| 262 | WebMall – A Multi-Shop Benchmark for Evaluating Web Agents Related Papers Related Patents Related Grants Related Venues Related Experts View Save Abstract: LLM-based web agents have the potential to automate long-running web tasks, such as searching for products in multiple e-shops and subsequently ordering the cheapest products that … |
Ralph Peeters; Aaron Steiner; Luca Schwarz; Julian Yuya Caspary; Christian Bizer; |
| 263 | Local Information Access in Marathi: Evaluating LLM-Native Web Retrieval in A Low-Resource Environment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce MarathiWeb, a benchmark for evaluating LLM-native web retrieval in Marathi that separates translated global queries from natively authored, local information needs. |
Sahil Kale; |
| 264 | IPQA: A Benchmark for Core Intent Identification in Personalized Question Answering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This gap is critical because without understanding which intents users prioritize, systems cannot generate responses satisfying individual information needs. To address this, we introduce the concept of core intents: intents users prioritize when selecting answers to satisfy their information needs. |
Jieyong Kim; Maryam Amirizaniani; Soojin Yoon; Dongha Lee; |
| 265 | Advantage-Conditioned Flow Policy for Offline Reinforcement Learning in Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose PerfRec (Preference-aware Flow for Recommendation), a flow-matching offline RL framework that learns an expressive behavioral policy and distills it into an efficient one-step policy. |
Xiaocong Chen; Siyu Wang; Lina Yao; |
| 266 | ReAlign: Optimizing The Visual Document Retriever with Reasoning-Guided Fine-Grained Alignment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose Reasoning-Guided Alignment (ReAlign), a method that enhances visual document retrieval by leveraging the reasoning capability of VLMs to provide fine-grained visual document descriptions as supervision signals for training. |
Hao Yang; Yifan Ji; Zhipeng Xu; Zhenghao Liu; Yukun Yan; Zulong Chen; Shuo Wang; Yu Gu; Ge Yu; |
| 267 | Follow The TRACE: Exploiting Post-Click Trajectories for Online Delayed Conversion Rate Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing methods address this through delay modeling or sample reweighting, yet neglect how post-click behaviors evolve over the observation period. To overcome this limitation, we formalize this evolution as feedback trajectory and propose TRACE. |
Xinyue Zhang; Yuanhao Ding; Xiang Ao; |
| 268 | PrepRet: Automated Data Preparation Pipeline Selection for Neural Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose PrepRet, a framework that jointly optimizes preprocessing pipeline selection and neural retrieval through differentiable optimization. |
Jing Chang; Chang Liu; |
| 269 | Unified Entity Matching Under Scarce Supervision Via Meta-Rule Induction and Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this regime, supervised unified models degrade substantially, and deployable compact LLMs remain unreliable: lightweight fine-tuning and in-context learning yield inconsistent behavior and can even exhibit negative effects under scenario shifts. To fill in this gap, we propose \o{}urs, a meta-rule induction and retrieval framework for unified entity matching under scarce supervision. |
Ziheng Zhang; Weixin Zeng; Jiuyang Tang; Xiang Zhao; |
| 270 | SocialDropout: Dynamic Agent Dropout for Social Simulation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Large language model driven multi-agent social simulation frameworks enable realistic modeling of complex societal dynamics but incur substantial computational overhead due to dense agent participation and extensive interaction costs. To address this limitation, we propose SocialDropout, a reinforcement learning–based agent selection strategy within the AgentSociety framework, inspired by the AgentDropout paradigm, which dynamically identifies and samples informative agent subsets for each simulation round. |
Huajie Wang; Geng Tu; Xi Zeng; Ruifeng Xu; Min Zhang; |
| 271 | RecGPT-Mobile: On-Device Large Language Models for User Intent Understanding in Taobao Feed Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose RecGPT-Mobile, a framework that designs a lightweight LLM-based intent understanding agent to improve recommendation quality in mobile e-commerce scenarios. |
Bin Zhang; Weipeng Huang; Dimin Wang; Jialin Zhu; Yuning Jiang; Zhaode Wang; Chengfei Lv; Jian Wang; Qichao Ma; Li Chen; Junqing Wu; Yipeng Yu; |
| 272 | Cross-Sensory Comparison of EEG Signals for Brain-Based Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Using the Brennan (auditory, 49 subjects) and Nieuwland (visual, 51 subjects) datasets, balanced for vocabulary overlap and sample distributions, we train BPR models with transformer EEG encoders and BERT text encoders via contrastive learning. |
Niall McGuire; Yashar Moshfeghi; |
| 273 | Mitigating Bias in Large Language Model Based Question Answering Through Causal Front Door Prompting Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose a Causal Front Door Prompting framework (CFDP) that reduces demographic influence by intervening on the chain of thought reasoning, which is treated as an observable mediator. |
Yaqi Yang; Ziqi Xu; Jie Li; Chenglong Ma; Jeffrey Chan; Mark Sanderson; Xin Zheng; Yongli Ren; |
| 274 | Beyond Unimodal Perspectives: Generative Retrieval with Multimodal Semantics Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Through a systematic analysis of Early and Late Fusion strategies, we reveal that naive integration fails due to two critical limitations: modality sensitivity, where one modality dominates the representation, and modality correspondence, where the model fails to align distinct semantic IDs across modalities. To overcome these challenges, we introduce MGR-LF++, an enhanced late fusion framework. |
Jing Zhu; Mingxuan Ju; Yozen Liu; Shubham Vij; Danai Koutra; Neil Shah; Tong Zhao; |
| 275 | PGSF: Profile-Guided Semantic Fusion for Decoupled Industrial Pre-Ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose Profile-Guided Semantic Fusion (PGSF), a plug-and-play framework that injects target-item guidance during training while preserving tower decoupling at inference. |
Chuike Sun; Nan Xu; Xing Fang; Yang Huang; Jing Wang; Junzhou Chen; |
| 276 | Generative Enhanced Modeling: A Collaborative Framework for Enhancing User Representations Via Semantic ID Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Furthermore, relying solely on a user’s own history limits exploration and reinforces the ”filter bubbles”. To address this, we propose GEM (Generative Enhanced Modeling). |
Li Li; Wei Xu; Yu Cheng; Jianbin Lin; Can Ye; |
| 277 | Uniboost: Global Coordination with Value Alignment for Fair and Efficient Traffic Allocation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing approaches often suffer from coupled allocation plans, score inflation, and a lack of interpretability. To address these challenges, we propose Uniboost, a unified traffic allocation framework. |
Ge Fan; Nan Zhao; Kai Meng; Cong Luo; Yang Fu; Huiping Chu; Jialin Liu; Yuning Jiang; Bo Zheng; |
| 278 | Learning to Trust: Dynamic Utilization of Retrieval-Augmented Generation for E-commerce Search Relevance Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose DyKnow-RAG, a reinforcement learning framework that teaches LLMs to learn to trust through dynamic utilization of external knowledge. |
Tingqiao Xu; Shaowei Yao; Chenhe Dong; Yiming Jin; Zerui Huang; Dan Ou; Haihong Tang; Bo Zheng; |
| 279 | Think When Needed: Model-Aware Reasoning Routing for LLM-based Ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Although reasoning prompting can enhance ranking utility, our preliminary exploration reveals that its benefits are inconsistent and come at a substantial computational cost, suggesting that when to reason is as crucial as how to reason. To address this issue, we propose a reasoning routing framework that employs a lightweight, plug-and-play router head to decide whether to use direct inference (Non-Think) or reasoning (Think) for each instance before generation. |
Huizhong Guo; Tianjun Wei; Dongxia Wang; Yingpeng Du; Ziyan Wang; Jie Zhang; Zhu Sun; |
| 280 | Gesture Clustering for Real-Time User Disentanglement in Shared-Account Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As a result, efficiently utilizing gesture information to provide more distinct identity signals for recommendation models remains a critical challenge. To address this issue, we propose G-CORE (Gesture Clustering for Real-time REcommendation), an unsupervised framework that disentangles gesture representations via clustering before integrating them into the main recommendation model. |
Huiying Hu; Xinlang Yue; Kexin Yi; Lingzhen Xu; Yangyi Fang; Muyang Li; Yongqi Liu; Kaiqiao Zhan; |
| 281 | Divergence Meets Consensus: A Multi-Source Negative Sampling Framework for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Negative sampling is significant for training sequential recommendation models under implicit feedback. The predominant strategy, self-guided hard negative sampling, selects negatives based on the model’s current state but suffers from three limitations: (1) the coupling between sampling and model updates triggers a vicious cycle that drives the model into local optima; (2) relying on current model parameters narrows sampling to a small region of the item space, reducing diversity and harming generalization; (3) identifying a hard negative requires scoring the entire candidate pool, causing substantial computational overhead with minimal information gain.To address these challenges, we propose MDCNS (Multi-source Divergence-Consensus for Negative Sampling), a novel ”Teacher-Peer-Self” framework inspired by Vygotsky’s Zone of Proximal Development (ZPD) theory. |
Yuanzi Li; Lingjie Wang; Jingyu Zhao; Zihang Tian; Yuhan Wang; Lei Wang; Xu Chen; |
| 282 | Depression Detection from Social Media: A Mutual Guidance Multi-modal Network with Complementary Graph Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing depression detection methods predominantly rely on static multi-modal fusion strategies and frequently fail to effectively tackle cross-modal semantic gaps. To address these limitations, we propose a Mutual Guidance Multi-modal Network with Complementary Graph Learning (MGMN) for depression detection by observing individuals’ behavioral performance on social media. |
Guocheng Hu; Chaoqun Zheng; Ruifan Zuo; Fengling Li; Dan Shi; Xiaofeng Qu; Wenpeng Lu; |
| 283 | MECI: Multi-Element Collaborative Interaction for Multimodal Entity Linking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In response, we propose the Multi-Element Collaborative Interaction (MECI) framework. |
Jie Peng; Yongxue Shan; Yongfu Zha; Xiaodong Wang; |
| 284 | Spectral-Adaptive Adversarial Hashing for Robust Image Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Adversarial training is the most effective method for improving robustness, but it often leads to a significant trade-off between robustness and retrieval accuracy. In this paper, we conduct spectral analysis and find that generating high-quality hash codes requires wide-frequency response models, whereas adversarial training forces the model into spectral collapse, degrading it to a low-frequency response model and weakening its discriminability. |
Gang Zhou; Shibiao Xu; Xiaolong Zheng; Daniel Dajun Zeng; |
| 285 | SIGMA: A Semantic-Grounded Instruction-Driven Generative Multi-Task Recommender at AliExpress Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Specifically, we first ground item entities in a unified latent space capturing both general semantics and collaborative signals. Building upon this, we introduce a hybrid item tokenization method for both precise modeling and efficient generation. |
Yang Yu; Lei Kou; Huaikuan Yi; Bin Chen; Yayu Cao; Lei Shen; Chao Zhang; Bing Wang; Xiaoyi Zeng; |
| 286 | Towards Unified Affective Reasoning Via In-Context Reinforcement Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing RL-based methods are largely confined to single-task settings, requiring separate training pipelines for different affective domains and limiting transfer across related problems. To address this limitation, we propose Unified Affective Reasoning (UAR), an in-context reinforcement learning framework for learning generalizable affective reasoning under a unified multi-task paradigm. |
Feng Xiong; Jun Wang; Ruifeng Xu; |
| 287 | PurifAI: Detecting and Fixing Search-Induced Distortions in Web-Augmented LLMs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While integrating real-time web search can enhance model utility, it introduces a critical vulnerability: the ingestion of conflicting, misleading, or hallucinated content from the open web can override the model’s adherence to its verified internal knowledge. We define this failure mode as search-induced distortion, a significant risk in high-stakes domains where the internal knowledge base serves as the absolute ground truth.To address this challenge, we present PurifAI, a proactive, model-agnostic, cache-level purification system designed for safety- and compliance-sensitive deployments. |
Guoqing Wang; Zhao Zhang; Zeyu Sun; Xiaofei Xie; Yizhou Chen; Yanchao Tan; Dan Hao; |
| 288 | L2Rec: Towards Dual-View Understanding of LLMs for Personalized Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we introduce L2Rec, which unifies behavioral and semantic understanding at the parameter level of LLMs. |
Pingjun Pan; Tingting Zhou; Peiyao Lu; Tingting Fei; Hongxiang Chen; Chuanjiang Luo; |
| 289 | Cross-Domain Preference Transfer for Promoting Engagement with Built-in Chatbots Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: It is non-trivial to transfer those cross-domain signals into meaningful suggested topics that are aligned with users’ real-time preference.In this paper, to promote user engagement with the built-in chatbot, we propose a cross-domain question generation framework via preference transfer, which consists of two stages: Cross-Domain Preference Adaptation and Target-Oriented Preference Alignment. |
Tianyu Wu; Guanyu Jiang; Hongwei Zheng; Haoming Li; Yongchun Zhu; Jingwu Chen; Feng Zhang; |
| 290 | The Attention Market: Interpreting Online Fair Re-ranking As Manifold Optimization Under Walrasian Equilibrium Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose ManifoldRank, an efficient online fair re-ranking algorithm. |
Chen Xu; Wei Chu; Wenyu Hu; Fengran Mo; Jun Xu; Maarten de Rijke; |
| 291 | NeuCon-ICE: Neuron-Level Controllable In-Context Editing for Multimodal Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Consequently, we propose a neuron-level controllable ICE framework for MKE, namely NeuCon-ICE. |
Chao Jiang; Jinzhi Liao; Xiang Zhao; |
| 292 | Where Relevance Emerges: A Layer-Wise Study of Internal Attention for Zero-Shot Re-Ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we conduct an orthogonal evaluation of generation, likelihood, and internal attention mechanisms across multiple ranking frameworks. |
Haodong Chen; Shengyao Zhuang; Zheng Yao; Guido Zuccon; Teerapong Leelanupab; |
| 293 | OntoEL: Neuro-Symbolic Biomedical Entity Linking with Differentiable Fuzzy EL⊥ Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present OntoEL, a neuro-symbolic framework that shifts BioEL from surface-level matching to logic-grounded reasoning. |
Chang Lu; Yizheng Zhao; |
| 294 | Decision-Theoretic Stopping Rules for Document Screening Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper applies decision theory to the problem and uses it to derive three practical stopping policies based on the Expected Value of Perfect Information. |
Aaron H.A. Fletcher; Mark Stevenson; |
| 295 | Confidence-Based Stopping Methods for Systematic Reviews Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Technology Assisted Review stopping methods aim to ensure that no more documents are screened than necessary. |
Aaron H.A. Fletcher; Mark Stevenson; |
| 296 | Information Seeking Behavior in LLM-Based RAG: Mental Models and Missing Information Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this Perspectives Paper, we argue that RAG interfaces compress the visible stages of the Information Search Process: exploration, comparison, and synthesis migrate inside the system, while the user sees only a conversational request-response loop. |
Bruno Nadalic Sotic; Jaap Kamps; |
| 297 | Who Is Shopping With You? How Persona Design Shapes Cognitive and Social Engagement in AI Shopping Agents Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address the gap, we conducted two studies in experience-goods domains. |
Hyungwoo Song; Kyusik Kim; Hyeonseok Jeon; Minjeong Shin; Bongwon Suh; |
| 298 | Meituan Merchant Business Diagnosis Via Policy-Guided Dual-Process User Simulation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Second, mechanism duality requires capturing both interpretable preferences and implicit statistical regularities, which no single paradigm achieves alone. We propose Policy-Guided Hybrid Simulation (PGHS), a dual-process framework that mines transferable decision policies from behavioral trajectories and uses them as a shared alignment layer. |
Ziyang Chen; Renbing Chen; Daowei Li; Jinzhi Liao; Jiashen Sun; Ke Zeng; Xiang Zhao; |
| 299 | Revisiting Collaborative Filtering By Unleashing The Power of Similarity Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Despite this, the expressiveness of high-performing similarity-based models is fundamentally confined to pairwise interactions within immediate neighborhoods, overlooking the combinatorial preference patterns in high-order connections. To overcome these limitations and fully unleash the potential of similarity-based models, we propose a novel bidirectional extension framework that systematically generalizes the modeling of similarity to group-level and multi-hop perspectives. |
Mingyang Li; Xinlang Yue; Chen Chen; Muyang Li; Yongqi Liu; Kaiqiao Zhan; |
| 300 | LWGR: Lagrangian-Constrained Personalized World Knowledge for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing methods rely on fixed, manually designed instructions to generate semantic knowledge and directly incorporate it into GR, which has two limitations: (1) fixed instructions cannot capture the multidimensional heterogeneity of user interests; (2) uncontrollable knowledge fusion may conflict with behavioral signals and harm recommendations. To address these limitations, we propose LWGR, a framework that leverages Lagrangian constraints to transfer users’ personalized World knowledge from LLMs into Generative Recommendation. |
Lingyu Mu; Hao Deng; Haibo Xing; Kaican Lin; Zhitong Zhu; Zhengxiao Liu; Zheng Lin; Xiaoyi Zeng; Yu Zhang; Jinxin Hu; |
| 301 | MIRA: An LLM-Assisted Benchmark for Multi-Category Integrated Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, current information retrieval (IR) evaluation benchmarks have not kept pace with this development, primarily due to the lack of test collections that represent the diversity of contemporary search domains. We address this critical gap with MIRA, a novel benchmark based on a large-scale social science search platform. |
Mehmet Deniz T\{u}rkmen; Suchana Datta; Dwaipayan Roy; Daniel Hienert; Philipp Mayr; Derek Greene; |
| 302 | RedGR: Unified Generative Retrieval for Recommendation in REDnote Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: (2) The mapping mechanism from SIDs to concrete items requires substantial refinement to ensure precise and reliable retrieval performance. To tackle these issues, we propose RedGR, a generative retrieval model that unifies the modeling of multiple complex retrieval tasks. |
Mengcheng Fang; Hongyu Wang; Xichuan Niu; Dong Xu; Honggang Wang; Tao Zhuang; Ping Yang; Yao Hu; |
| 303 | WeSEAL: Well-calibrated Search for Eliminating Attention-sink Leakage Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: By employing a contrastive head selection mechanism and a novel attention-aligned Supervised Fine-Tuning objective, WeSEAL suppresses background noise and redirects discriminative signals to semantic retrieval heads, thereby intrinsically correcting positional bias. |
Juyuan Wang; Chenxing Wang; Aolin Li; Huiyun Hu; Yuchen Fang; Haijun Wu; Jin Xu; Dongliang Liao; |
| 304 | LiBRA: Enhancing Live-streaming Recommendation with Bigraph-based Capacity-aware Ranking Approach Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper formulates the viewer-streamer match problem as a capacity-aware optimal subgraph problem on the viewer-streamer bipartite graph, and proposes a novel capacity-aware ranking approach for live-streaming recommendation systems with explicit consideration of streamers’ limited service capacity. |
Youran Zhang; Xiaotong Guo; Jiayi Lu; Runze Yu; Zhuang Cai; Qinglei Wang; |
| 305 | Next Interest Flow: A Generative Pre-training Paradigm for Recommender Systems By Modeling All-domain Movelines Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present the All-domain Moveline Evolution Network (AMEN), a unified framework implementing this pipeline. |
Chen Gao; Zixin Zhao; Lv Shao; Tong Liu; |
| 306 | Efficient Sparse Retrieval with Lightweight Superblock Pruning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper proposes a simple and effective superblock pruning scheme that reduces the overhead of superblock score computation while preserving competitive relevance. |
Parker Carlson; Wentai Xie; Rohil Shah; Tao Yang; |
| 307 | Scalable K-Means Guided Partitioning for Block-based Sparse Document Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: The K-means method has been widely adopted to group similar sparse vectors together, but it is not scalable when dealing with a large number of clusters, and the bipartite graph partitioning (BP) method is a preferred choice for block-based document retrieval to partition a large document set efficiently. This paper revisits such a partitioning approach and proposes a balanced K-means-guided bisection method with log-linear complexity while maintaining a good similarity-based clustering quality. |
Parker Carlson; Sammy Lesner; Antonio Mallia; Tao Yang; |
| 308 | Bridging Passive and Active: Enhancing Conversation Starter Recommendation Via Active Expression Modeling Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As a result, the system skews towards popular but generic suggestions. In this work, we uncover an untapped paradigm shift to shatter this harmful feedback loop: harnessing user ”free will” through active user expressions. |
Yiqing Wu; Haoming Li; Guanyu Jiang; Jiahao Liang; Yongchun Zhu; Jingwu Chen; Feng Zhang; |
| 309 | Beyond Hard Negatives: The Importance of Score Distribution in Knowledge Distillation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we highlight that focusing solely on hard negatives prevents the student from learning the comprehensive preference structure of the teacher, potentially hampering generalization. |
Youngjoon Jang; Seongtae Hong; Hyeonseok Moon; Heuiseok Lim; |
| 310 | FairSpec: Expert Specialization for Fair LLM-based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While fine-tuning approaches show promise, improving fairness typically comes at the cost of degraded recommendation performance, as the model’s parameters conflates user preferences with sensitive attributes. To address this, we propose FairSpec, a novel lightweight fine-tuning framework designed to enhance fairness while preserving recommendation quality. |
Yuchen Zheng; Xuan Pan; Jing Wang; Chuanchang Zhang; Xi Lin; Chunyao Song; Xiangrui Cai; Xiaojie Yuan; |
| 311 | Refairmulate: A Large-Scale Dataset for Gender-Fair Query Reformulations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce Refairmulate, an open resource that supports training and benchmarking gender-fair query reformulation models through a multi-objective procedure that jointly balances retrieval effectiveness and gender bias. |
Hai Son Le; Shirin Seyedsalehi; Morteza Zihayat; Ebrahim Bagheri; |
| 312 | GNN-based Anchor Embedding for Efficient Subgraph Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Several recent works utilize deep learning (DL) techniques for subgraph retrieval via matching, yet most only return approximate isomorphism relations between queries and data graphs–failing to retrieve all exact matching locations, a critical demand for structured graph retrieval in information retrieval. Unlike these DL-based approximate methods, we propose a learning-based framework for subgraph retrieval, called the graph neural network (GNN)-based anchor embedding framework (GNN-AE), which can efficiently retrieve all exact matching locations. |
Bin Yang; Jianxiong Ye; Zhaonian Zou; |
| 313 | Balanced Frequency Decoupling: Energy-Aware Multi-Scale Preference Modeling for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Moreover, they generally rely on a coupled modeling mechanism that handles both long-term and short-term preferences within a single backbone network, lacking dedicated modeling paths tailored to their distinct temporal characteristics. To address these challenges, we propose a Balanced Frequency Decoupling Sequential Recommendation model (BFDRec). |
Jiahao Hu; Wei Zhou; Jie Liao; Junlin Zhu; Junhao Wen; Hongyu Zhang; |
| 314 | ARHN: Answer-Centric Relabeling of Hard Negatives with Open-Source LLMs for Dense Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: By relying strictly on open-source models, ARHN establishes a cost-effective and scalable refinement pipeline suitable for large-scale training. |
Hyewon Choi; Jooyoung Choi; Hansol Jang; Hyun Kim; Chulmin Yun; Changwook Jun; Stanley Jungkyu Choi; |
| 315 | FactOWL: A Cost-Efficient Tool for Long-Form Factuality Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing implementations suffer from slow inference, outdated knowledge bases, and the use of paid search and LLM APIs, which hinders research and practical applications. To fill this gap, we propose FactOWL, a FActScore-based Factuality evaluation tool which adopts an Open LLM and real-time Wikipedia search for evaluation of Long-form LLM responses. |
Andrey Sakhovskiy; Nikita Sushko; Maria Marina; Vasily Konovalov; Elena Tutubalina; Alexander Panchenko; Pavel Braslavski; |
| 316 | FollowTable: A Benchmark for Instruction-Following Table Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To support systematic evaluation, we introduce FollowTable, the first large-scale benchmark for IFTR, constructed via a taxonomy-driven annotation pipeline. |
Rihui Jin; Yuchen Lu; Ting Zhang; Jun Wang; Kuicai Dong; Zhaocheng Du; Dongping Liu; Gang Wang; Yong Liu; Guilin Qi; |
| 317 | SOLARIS:Speculative Offloading of Latent-bAsed Representation for Inference Scaling Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While these models offer superior performance, their computational demands make real-time serving impractical, often forcing practitioners to rely on knowledge distillation—compromising serving quality for efficiency. To address this challenge, we present SOLARIS (Speculative Offloading of Latent-bAsed Representation for Inference Scaling), a novel framework inspired by speculative decoding. |
Zikun Liu; Liang Luo; Qianru Li; Zhengyu Zhang; Wei Ling; Jingyi Shen; Zeliang Chen; Yaning Huang; Jingxian Huang; Abdallah Aboelela; Chonglin Sun; Feifan Gu; Fenggang Wu; Hang Qu; Huayu Li; Jill Pan; Kaidi Pei; Laming Chen; Longhao Jin; Qin Huang; Tongyi Tang; Varna Puvvada; Wenlin Chen; Xiaohan Wei; Xu Cao; Yantao Yao; Yuan Jin; Yunchen Pu; Yuxin Chen; Zijian Shen; Zhengkai Zhang; Dong Liang; Ellie Wen; |
| 318 | Learning Evidence of Depression Symptoms Via Prompt Induction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This task is fine-grained and highly imbalanced, and we find that common LLM approaches (zero-shot, in-context learning, and fine-tuning) struggle to apply consistent relevance criteria for most symptoms. We propose Symptom Induction (SI), a novel approach which compresses labeled examples into short, interpretable guidelines that specify what counts as evidence for each symptom and uses these guidelines to condition classification. |
Eliseo Bao; Anxo Perez; David Otero; Javier Parapar; |
| 319 | LLM-based Semantic and ID Representations for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Although prior works have made significant progress, there are still three key challenges: (i) how to leverage the open-world knowledge and reasoning ability of LLM to enhance item representations; (ii) how to adaptively capture semantic preferences and collaborative preferences from multi-modality sequences; and (iii) how to construct signals to optimize model training and guide the fusion of ID and semantic information. To address these challenges, we propose a novel model, i.e., LLM-based semantic and ID representations for sequential recommendation (SIDSRec). |
Donglin Zhou; Weike Pan; Zhong Ming; |
| 320 | DIAURec: Dual-Intent Space Representation Optimization for Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This imbalance has led to limited progress, as representation optimization is crucial for recommendation quality by promoting the affinity between users and their interacted items in the feature space, yet remains largely overlooked. To overcome these limitations, we propose DIAURec, a novel representation learning framework that unifies intent and language modeling for recommendation. |
Yu Zhang; Yiwen Zhang; Yi Zhang; Lei Sang; |
| 321 | IR Lens: A Tool for Interpreting Cross-Encoder Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This hinders not only our understanding of the systems implementing them, but also our ability to improve them. To alleviate this limitation, we introduce IR Lens, a new interpretability tool tailored to cross-encoders based on two key components: 1) Neuron Integrated Gradients to expose the contributions of model parts at multiple levels, and 2) targeted ablations to support hypothesis tracking. |
Mihai Branga-Peicu; Mathias Vast; Basile Van Cooten; Laure Soulier; Jules Fran\c{c}oise; Benjamin Piwowarski; Baptiste Caramiaux; |
| 322 | Breaking The Relevance–Diversity Seesaw: Hierarchical LLM Reasoning with RL for Industrial Novelty Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present BALANCE, a hierarchical reasoning-and-generation framework that decomposes novelty recommendation into three structured stages: generating a Novelty Tag for exploration direction, refining an Interest Topic for intent specification, and constructing a Recommendation List for facet coverage. |
Ying Sun; Yanyan Zou; Xiao Wang; Hanchuan Xu; Xuanhua Yang; Sulong Xu; Junbo Qi; Shengjie Li; |
| 323 | Hierarchical Denoising Entire Space Multi-Task Model for Post-Click Conversion Rate Prediction with Noisy Labels Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Crucially, within the entire space modeling framework, this noise exhibits a complex hierarchical structure, where the superimposition of noise from antecedent click labels and subsequent conversion labels results in conventional single-task denoising methods being ineffective. To bridge this gap, we propose HiDe-ESMM (Hierarchical Denoising Entire Space Multi-Task Model), a unified framework designed to systematically mitigate hierarchical label noise. |
Yan Lyu; Haoxuan Li; Tianyu Xia; Xiang Li; Xiangnan Feng; Chunyuan Zheng; Xiao-Hua Zhou; |
| 324 | RAD-DPO: Robust Adaptive Denoising Direct Preference Optimization for Generative Retrieval in E-commerce Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While Direct Preference Optimization (DPO) offers an efficient alignment solution, its direct application to structured SIDs suffers from three limitations: (i) it penalizes shared hierarchical prefixes, causing gradient conflicts; (ii) it is vulnerable to noisy pseudo-negatives from implicit feedback; and (iii) in multi-label queries with multiple relevant items, it exacerbates a probability squeezing effect among valid candidates. To address these issues, we propose RAD-DPO, which introduces token-level gradient detachment to protect prefix structures, similarity-based dynamic reward weighting to mitigate label noise, and a multi-label global contrastive objective integrated with global SFT loss to explicitly expand positive coverage. |
Zhiguo Chen; Guohao Sun; Yiming Qiu; Xingzhi Yao; Mingming Li; Huimu Wang; Yangqi Zhang; Songlin Wang; Sulong Xu; |
| 325 | BEATS: Bootstrapping E-commerce Attribute Taxonomies for Search Through Iterative Human-AI Collaboration Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present BEATS, a human-in-the-loop LLM framework for bootstrapping product attribute taxonomies entirely from scratch. |
Yung-Yu Shih; Shang-Yu Su; Tzu-I Ho; Dongzhe Wang; Yun-Nung Chen; |
| 326 | Enhancing Multi-Valued Treatment Uplift Modeling with Knowledge Sharing and RCT Data Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose CAMU, which combines (i) a Codebook-enhanced representation network that anchors representations across mini-batches to stabilize learning and facilitate distribution alignment, and (ii) a Cross-Treatment Self-Attention (CTSA) enhanced prediction network for treatment-aware knowledge sharing. |
Jiaxiang Liu; Luo He; Zhenghao Zeng; Deqiang Kong; Funan Mu; Jiaming Zhang; Zilong Lu; Peng Jiang; Hongyan Liu; |
| 327 | Adaptive Autoguidance for Item-Side Fairness in Diffusion Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Diffusion recommender systems achieve strong recommendation accuracy but often suffer from popularity bias, resulting in unequal item exposure. To address this shortcoming, we introduce A2G-DiffRec, a diffusion recommender that incorporates adaptive autoguidance, where the main model is guided by a less-trained version of itself. |
Zihan Li; Gustavo Escobedo; Marta Moscati; Oleg Lesota; Markus Schedl; |
| 328 | Corpus-Centric Learning for Zero-Shot Table Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose GeCo-TR (Generative Schema and Contrastive Table Retrieval), a zero-shot table retrieval framework that eliminates the need for supervised QA data by shifting from direct query-to-table learning to modeling the intrinsic structural semantics of the table corpus. |
Zhou He; Zhifei Pang; Xiu Tang; Sai Wu; Gang Chen; |
| 329 | CoMCo: Consistency-Aware Multi-Agent Coordination for Zero-Shot Cross-Modal Entity Matching Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Multimodal large language models (MLLMs) offer richer cross-modal cues for matching, but exhaustive MLLM reasoning over large candidate spaces is prohibitively expensive. To address these limitations, in this work, we propose \o{}urq, a blackboard-based multi-agent framework for zero-shot cross-modal entity matching that performs iterative, self-correcting refinement by fusing multiple matching signals, explicitly regulating global consistency, and selectively invoking an MLLM only for hard cases. |
Shiqi Zhang; Weixin Zeng; Ziheng Zhang; Wenzhe Hou; Weidong Xiao; Xiang Zhao; |
| 330 | Drift-Aware Incremental Token Adaptation with Collaborative Semantics for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To balance plasticity and stability for collaborative tokenizers, we propose DACT, a Drift-Aware Continual Tokenization framework with two stages: (i) tokenizer fine-tuning, augmented with a jointly trained Collaborative Drift Identification Module (CDIM) that outputs item-level drift confidence and enables differentiated optimization for drifting and stationary items; and (ii) hierarchical code reassignment using a relaxed-to-strict strategy to update token sequences while limiting unnecessary changes. |
Yuebo Feng; Jiahao Liu; Mingzhe Han; Dongsheng Li; Hansu Gu; Peng Zhang; Tun Lu; Ning Gu; |
| 331 | RankEvolve: Automating The Discovery of Retrieval Algorithms Via LLM-Driven Evolution Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce RankEvolve, a program evolution setup based on AlphaEvolve, in which candidate ranking algorithms are represented as executable code and iteratively mutated, recombined, and selected based on retrieval performance across 12 IR datasets from BEIR and BRIGHT. |
Jinming Nian; Fangchen Li; Dae Hoon Park; Yi Fang; |
| 332 | Learning to Rank with Multi-Criteria LLM-Judge Annotations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We evaluate this approach using manual relevance labels from TREC TREC DL 2019, DL 2020, and DL 2023. |
Naghmeh Farzi; Laura Dietz; |
| 333 | Auto-Judge: A Cross-Task Benchmark for Comparing LLM Judges for Citation-Grounded RAG Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present the Auto-Judge resource for the meta-evaluation of automated LLM judges, especially judges that evaluate Retrieval-Augmented Generation (RAG) systems that ground their response with citations. |
Naghmeh Farzi; Tim Hagen; Eugene Yang; Maik Fr\{o}be; Ronak Pradeep; Hossein A. Rahmani; Xi Wang; Oleg Zendel; Martin Potthast; Laura Dietz; |
| 334 | Too Many Questions: Deriving Concise and Effective Nugget Banks Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Inspired by preference-based evaluation, we derive differential nuggets from winner-loser passage pairs, focusing on information that captures differences in topicality, level of detail, and evidential support between responses under an automatic preference judge. |
Laura Dietz; Naghmeh Farzi; Eugene Yang; Dawn Lawrie; |
| 335 | When \& How to Write for Personalized Demand-aware Query Rewriting in Video Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, traditional methods utilizing implicit history features often suffer from signal dilution and delayed feedback. To address these challenges, we propose WeWrite, a novel Personalized Demand-aware Query Rewriting framework. |
Cheng Cheng; Chenxing Wang; Aolin Li; Haijun Wu; Huiyun Hu; Juyuan Wang; Dongliang Liao; |
| 336 | Reconstructing Content with Collaborative Attention for Universal Multimodal Representation Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While effective for generation, the causal attention and next-token prediction paradigm of MLLMs does not explicitly encourage the formation of globally compact representations, limiting their effectiveness as multimodal embedding backbones. To address this, we propose CoCoA, a Content reconstruction pre-training paradigm based on Collaborative Attention for universal multimodal representation learning. |
Jiahan Chen; Da Li; Hengran Zhang; Yinqiong Cai; Lixin Su; Jiafeng Guo; Daiting Shi; Dawei Yin; Keping Bi; |
| 337 | PatentMerit: A Holistic System for In-depth Technology Understanding Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Hence, we develop PatentMerit, a comprehensive system for in-depth analysis of the value of patents, based on large language models, network analysis, and data mining techniques. |
Tianxiang Xie; Jingxuan Wu; Jiaying Liu; Junxiang Zhang; Shuo Yu; |
| 338 | Not All Imputations Are Trustworthy: An Uncertainty-aware Multi-modal Entity Alignment Framework Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Ignoring the aleatoric uncertainty of such modalities introduces severe noise which propagates through the fusion process and degrades alignment performance. To address this challenge, we propose a novel MMEA framework, namely SURE, to Suppress Uncertainty for tRustworthy Entity alignment. |
Weijie Wang; Shijie Luo; Xinyuan Lu; Qinpei Zhao; Weixiong Rao; |
| 339 | Towards Conflict-aware Selective Knowledge Unlearning for Continual Few-shot Knowledge Graph Completion Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To bridge this research gap, we propose a novel FKGC framework named CSKU, which achieves dynamic knowledge adaptation by selectively unlearning structural and semantic conflicts. |
Junlin Zhu; Bo Fu; Guiduo Duan; |
| 340 | R&F-Inventory: A Large-Scale Dataset for Monotonic Inventory Estimation in Reach and Frequency Advertising Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We further derive the theoretical maximum exposure ceiling and use it as a consistency check to evaluate data quality and the feasibility of model predictions. Using this data set, this paper defines two standardized benchmark tasks: single-point performance prediction and reconstruction of budget-performance curves, and provides a set of reproducible baseline methods and evaluation protocols. |
Yunshan Peng; Ji Wu; Wentao Bai; Yunke Bai; Jinan Pang; Wenzheng Shu; Yanxiang Zeng; Xialong Liu; Peng Jiang; |
| 341 | SRAG: A Lightweight and Specialized Retrieval-augmented Generation System at The Edge Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Through extensive empirical analysis, we find that domain-specialized knowledge bases, when deployed on individual edge servers, deliver substantially higher retrieval accuracy and generation quality than generic knowledge bases under identical resource budgets. Based on this, we propose SRAG, a distributed RAG system that enforces knowledge specialization at the edge. |
Ruikun Luo; Zihan Xing; Lin Gu; Song Wu; Hai Jin; Xiaoyu Xia; |
| 342 | A Sketch+Text Composed Image Retrieval Dataset for Thangka Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we introduce CIRThan, a sketch+text composed image retrieval dataset for Thangka imagery, a culturally grounded and knowledge-specific visual domain characterized by complex structures, dense symbolic elements, and domain-dependent semantic conventions. |
Jinyu Xu; Yi Sun; Jiangling Zhang; Qing Xie; Daomin Ji; Zhifeng Bao; Jiachen Li; Yanchun Ma; Yongjian Liu; |
| 343 | A General Framework for Multimodal LLM-Based Multimedia Understanding in Large-Scale Recommendation Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While Multimodal Large Language Models (MM-LLMs) offer robust mechanisms for interpreting such complex data, their integration into latency-constrained, industrial-scale architectures remains a significant challenge. To address this, we propose a generalized framework for MM-LLM-driven multimedia understanding. |
Yiming Zhu; Xu Liu; Ziyun Xu; Zheng Wu; Joena Zhang; Sirius Chen; Chenheli Hua; Silvester Yao; Qichao Que; Wentao Shi; Junfeng Pan; Linhong Zhu; |
| 344 | When Vision Meets Texts in Listwise Reranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose Rank-Nexus, a multimodal image-text document reranker that performs listwise qualitative reranking on retrieved lists incorporating both images and texts. |
Hongyi Cai; |
| 345 | CATS: Cluster-Aware Thompson Sampling for Negative Mining in Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose CATS (Cluster-Aware Thompson Sampling),an adaptive negative sampling method that balances exploration and exploitation by leveraging unsupervised item clustering and adaptive multi-armed bandit principles. |
Giulia Di Teodoro; Federico Siciliano; Nicola Tonellotto; Fabrizio Silvestri; |
| 346 | Incremental Multi-Behavior Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we propose a novel solution called Behavior-aware Incremental Graph Convolution Network with Multi-Task Learning (BIGCN-MTL) for IMBR. |
Jiahao Gong; Weike Pan; |
| 347 | Debiased Recommendation Beyond The Positive Propensity Assumption Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this study, we theoretically show that when such zero-propensity samples, termed extrapolation samples exist, both IPS and DR estimators become biased. To overcome this limitation, we propose ExtraDebias method, which enables debiased recommendation in both non-extrapolation and extrapolation samples. |
Yanghao Xiao; Hao Wang; Xiang Li; Qian Zou; Cheng Bing; Wei Lin; Haoxuan Li; Zhouchen Lin; |
| 348 | LiveRAG: A Diverse Q&A Dataset with Varying Difficulty Level for RAG Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce the LiveRAG benchmark, a publicly available dataset of 895 synthetic questions and answers designed to support systematic evaluation of RAG-based Q&A systems. |
David Carmel; Simone Filice; Guy Horowitz; Yoelle Maarek; Alexander Shtoff; Oren Somekh; Ran Tavory; |
| 349 | The Matryoshka Hypencoder Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Inspired by Matryoshka Representation Learning, we show that the Hypencoder can be extended to support multiple sizes of Q-Nets, allowing trade-offs between effectiveness and efficiency when deployed. |
Majd Alkawaas; Sean MacAvaney; |
| 350 | ZoRRO: A Zero-Weight Personalized Recommender System for Scalable News Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present ZoRRO (Zero-Weight Personalized Recommender System), a zero-weight and training-free framework for personalized news recommendation designed for scalable real-world deployment. |
Johannes Kruse; Ryotaro Shimizu; Kasper Lindskow; Jon Tofteskov; Michael Riis Andersen; Julian McAuley; Jes Frellsen; |
| 351 | XFoodRec: An Explainable Mobile Recommender for Personalized Healthy Eating Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present XFoodRec, an extensible mobile research platform designed to bridge the gap between offline algorithmic evaluation and online behavioral studies. |
Amir Mollazadeh; Mourad Oussalah; Mehrdad Rostami; |
| 352 | Answer First, Evidence Second? Uncovering Hidden Risks in Well-Structured AI Search Summaries Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: These summaries are typically well-structured and citation-rich, creating a strong appearance of reliability that can encourage user trust, even when evidential grounding is uncertain. Motivated by this gap between appearance and evidence, we analyze 14,175 real-world queries from MS MARCO by examining Google Search AI summaries and their cited sources. |
Jinman Li; Xuanang Chen; Ruoxi Xu; Hongyu Lin; Yaojie Lu; Zecheng Fan; Xianpei Han; Le Sun; |
| 353 | ACE: Semantically-Grounded Graph Alignment Via Affective Contrastive Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose ACE (Affective Contrastive Embeddings), a framework that improves representation alignment by incorporating affective semantics into contrastive learning. |
Potito Aghilar; Sabino Roccotelli; Vito Walter Anelli; Alejandro Bellogin; Michelantonio Trizio; Tommaso Di Noia; |
| 354 | F2-Gen: An Open-Source Web Platform for Scenario-Driven Financial Fraud Data Simulation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Abstract: Real bank-transaction data are rarely available to non-regulatory parties due to privacy and compliance constraints, while public substitutes often mismatch financial semantics, … |
Junquan Gu; Zehao Gong; Runchen Ji; Kun Shi; Xiangfeng Luo; Hang Yu; |
| 355 | WPGRec: Wavelet Packet Guided Graph Enhanced Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In addition, collaborative signals from the user-item interaction graph are often injected through scale-inconsistent auxiliary modules, making cross-scale interference difficult to control. To address these issues, we propose WPGRec, a unified time-frequency and graph-enhanced framework that aligns multi-resolution temporal modeling with graph propagation at matching scales. |
Peilin Liu; Zhiquan Ji; Gang Yan; |
| 356 | Same Frequency Begets Shared Interests: Popular-Niche Wavelet Graph Learning for Multimodal Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This deficiency leads to problems such as the information of popular items overshadowing that of niche items during message passing and multimodal fusion, thereby compromising the accuracy and diversity of recommendations. To address these problems, we propose a novel Popular-Niche Graph Wavelet Learning Framework for Multimodal Recommendation (PNGRec). |
Yue He; Hongbo Chen; Jingxi Xie; Fengling Li; Jingjing Li; |
| 357 | CMSL: Constructive Multi-Sequence Learning for Recommendation Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This ”noisy” signal dilutes the model’s focus, effectively capping its ability to discern high-intent patterns from background activity. To address this, we propose Constructive Multi-Sequence Learning (CMSL), a paradigm shift from passive sequence ingestion to active ”context engineering” that constructs multiple coherent sequences in latent space. |
Zikun Cui; Renzhi Wu; Junjie Yang; Li Sheng; Jijie Wei; Linfeng Liu; Tai Guo; Tao Jia; Xiaodong Wang; Hong Li; Li Yu; Sri Reddy; Hong Yan; |
| 358 | How Generative AI Disrupts Search: An Empirical Study of Google Search, Gemini, and AI Overviews Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we aim to understand how generative AI disrupts web search by retrieving and presenting the information and sources differently from traditional search engines. |
Riley Grossman; Songjiang Liu; Michael K. Chen; Mike Smith; Cristian Borcea; Yi Chen; |
| 359 | Seeing The Whole Through The Parts: Discovering Objects Through Semantic Part Mining in Weak Supervision Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Building on this perspective, we propose P2WDet (Part-to-Whole Detection), a novel framework that shifts WSOD from traditional instance selection to semantic reconstruction. |
Shucheng Li; Weixuan Xu; Le Jiang; Hao Wu; Fengyuan Xu; Fan Wu; Feng Lyu; |
| 360 | ACE: Anisotropy-Controllable Embedding for LLM-enhanced Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Most vectors are concentrated in similar directions, resulting in a geometric imbalance that makes it difficult to adapt to collaborative signals during fine-tuning. To address this challenge, we propose Anisotropy-Controllable Embedding (ACE), which explicitly controls the anisotropy of LLM-generated embeddings. |
Dongcheol Lee; Hye-young Kim; Jongwuk Lee; |
| 361 | MonacGraph: A Monadic Second-Order Logic Extended Graph Database System with Community-Aware Storage Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present Monac- Graph, a graph database system that enables practical MSOL queries. |
Yuntao Jin; Songyao Wang; Chaokun Wang; |
| 362 | DCGL: Dual-Channel Graph Learning with Large Language Models for Knowledge-Aware Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Nevertheless, current KG&LLM-based methods still face three main limitations: 1) inadequate modeling of implicit semantics relationships beyond explicit KG links; 2) suboptimal single-channel fusion of ID and LLM embeddings, which often leads to signal interference and blurred representations; and 3) insufficient consideration of user-item interaction frequency variations in recommendation strategies. To address these challenges, we propose the Dual-Channel Graph Learning (DCGL) framework, featuring three key innovations: 1) a dual-channel architecture that structurally decoupling rich semantic information from user behavioral patterns, preventing early interference; 2) a multi-level contrastive learning mechanism that enhances robustness against KG noise through intra-view contrast and bridges semantic gaps between channels via inter-view alignment; and 3) a dynamic fusion mechanism that adaptively balances semantic generalization and behavioral specificity based on interaction frequency, resolving the cascading limitation. |
Xinchi Zou; Tongzhenzhi Su; Jianjun Li; Yuan Fu; Chang Liu; Zhiying Deng; Zhiwei Shen; |
| 363 | An Action-Aware Generative Sequence Modeling for Short Video Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Considering that user consumption is a temporal process, this paper demonstrates that the timing of user actions can represent diverse intentions through statistical analysis and examination of action patterns. Based on this insight, we propose a novel modeling paradigm: Action-Aware Gen erative Sequence Network (A2Gen ), which refines user actions (e.g., Like and Follow, etc.) along the temporal dimension and chains them into sequences for unified processing and prediction. |
Wenhao Li; Zihan Lin; ZhengXiao Guo; Jie Zhou; Shukai Liu; Yongqi Liu; Chuan Luo; |
| 364 | Probe-then-Plan: Environment-Aware Planning for Industrial E-commerce Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While large language models (LLMs) offer powerful reasoning capabilities, existing LLM-based search paradigms suffer from a fundamental blindness-latency dilemma: query rewriting methods are agnostic to retrieval tool capabilities and real-time inventory states, resulting in invalid plans; conversely, deep search agent approaches initially plan without environment awareness, then rely on iterative tool calls and reflection to perceive and correct failures, leading to seconds of latency, incompatible with the sub-second budget for the planning module in industrial e-commerce search. To resolve this conflict, we propose Environment-Aware Search Planning (EASP), a novel paradigm that reformulates search planning as a dynamic reasoning process grounded in environmental reality. |
Mengxiang Chen; Zhouwei Zhai; Jin Li; |
| 365 | DPEO: Dynamic Preference Evolution Optimization for Self-Evolving CTR Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose DPEO (Dynamic Preference Evolution Optimization), a co-evolutionary framework that transforms CTR modeling into a dynamic policy contention task. |
Kun Yao; Congcong Liu; Ziheng Ni; Wenlong Chen; Changping Peng; Ching Law; |
| 366 | AIPO: Adaptive Anchored Intent-aware Policy Optimization for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Massive click noise and sparse rewards result in highly unstable policy gradients, making it difficult to balance reward optimization and generative stability. To address this, we propose AIPO (Adaptive Anchored Intent-aware Policy Optimization) framework. |
Kun Yao; Congcong Liu; Ziheng Ni; Cai Shang; Wenlong Chen; Changping Peng; Ching Law; |
| 367 | Diagnosing and Mitigating Mid-Sequence Degradation in Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Through controlled perturbations, we find that this U-shaped bias persists even when collaborative signals are moved, indicating an architectural effect rather than a positional-encoding artifact. To mitigate this bias, we propose Disentangling and Calibrating Position Bias framework, which is a lightweight, plug-and-play calibration procedure that combines bias-adaptive masking during training with counterfactual logit calibration at inference to remove residual position-only effects. |
Linjiang Guo; Nitin Bisht; Shiqing Wu; Huan Huo; Xianzhi Wang; Guandong Xu; |
| 368 | Inductive Subgraphs As Shortcuts: Causal Disentanglement for Heterophilic Graph Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, the fundamental reasons behind misclassifications remain poorly understood. In this work, we take a novel perspective by examining recurring inductive subgraphs, empirically and theoretically showing that they act as spurious shortcuts that mislead GNNs and reinforce non-causal correlations in heterophilic graphs. |
Xiangmeng Wang; Qian Li; Haiyang Xia; Hao Miao; Qing Li; Guandong Xu; |
| 369 | Partial Label Learning-Inspired Denoising Implicit Feedback for Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Specifically, we propose PLLD, a Partial Label Learning-inspired Denoising method. |
Huilin Chen; Jie Lu; Kezhi Lu; Zhen Fang; Guangquan Zhang; |
| 370 | When Context Bites: Detecting RAG Poisoning Via Document-Level Attention Collapse Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Unlike the dispersed attention in benign generations, attacked generations exhibit a decrease in entropy as attention concentrates on poisoned documents. Building on these findings, we propose D-SCAN (Document-level Signal Collapse Analysis), a lightweight detection framework that monitors attention dynamics to identify attacked generations. |
Yingtao Ren; Ziyi Zhao; Yiwei Fu; Xiao Luo; Yu-Cheng Chang; Chin-Teng Lin; |
| 371 | Privacy Preserving Information Retrieval: Defining Privacy Research Pillars for A Future Research Agenda Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: How has the privacy threats landscape changed, and in which directions should the IR and Privacy research community investigate to address such new risks? In this perspective paper, we provide initial answers to these questions, analysing state-of-the-art solutions for protecting user privacy when accessing information and highlighting areas of concern. |
Francesco Luigi De Faveri; Guglielmo Faggioli; Asia J. Biega; Nicola Ferro; |
| 372 | Looking for The Bottleneck in Fine-grained Temporal Relation Classification Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we revisit the broader goal of classifying interval relations between temporal entities by considering the full set of relations that can hold between two time intervals. |
Hugo Sousa; Ricardo Campos; Alipio Jorge; |
| 373 | Multi-Faceted Continual Knowledge Graph Embedding for Semantic-Aware Link Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: A shared embedding fails to capture and distinguish these temporal semantic variations, degrading lifelong link prediction accuracy across snapshots. To address this, we propose a Multi-Faceted CKGE framework (MF-CKGE) for semantic-aware link prediction. |
Jing Qi; Yuxiang Wang; Zhiyuan Yu; Xiaoliang Xu; Yuanshi Zheng; Tianxing Wu; |
| 374 | Exploration-and-Thinking: Agentic Reasoning Over Knowledge Graphs Via An LLM-RL Synergized Framework Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To this end, we propose EAT (Exploration-and-Thinking), a novel agentic framework that synergizes LLMs with Reinforcement Learning (RL) for efficient and faithful reasoning on KGs. |
Yi Xia; Gang Zhou; Jing Chen; Xiaohui Chen; Qinlong Fan; Shunhang Li; |
| 375 | Learning to Forget: Satiation-Aware Long-Sequence Transducers for Mitigating Post-Purchase Redundancy Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose the Satiation-Aware Mechanism (SAM), an end-to-end framework designed to explicitly model the lifecycle of user interests. |
Yipin Dai; Ruocong Tang; Xing Fang; Yang Huang; Jing Wang; Zhentao Song; He Guo; |
| 376 | Unifying Granularity and Reliability: A Robust and Efficient Framework for Text-based Person Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, applying PETL to TPR remains challenging, as its limited adaptation capacity struggles to capture intricate identity cues and becomes highly susceptible to gradient interference from unreliable image-text pairs. To address these challenges, we present a PETL-based framework named UniGR that unifies granularity and reliability for robust and efficient TPR. |
Jingchen Hao; Jiang Liu; Zhen Peng; Yuting Zhang; Zhongjiang He; Weizhan Zhang; Hao Sun; |
| 377 | Don’t Contrast The Impossible: Region-Constrained Batching for Contrastive User Modeling on A Local Community Platform Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: On local community platforms such as Karrot, however, exposure is geographically constrained; many user-item pairs are impossible by design yet still treated as negatives during training, diluting the contrastive learning signal. We address this impossible negatives problem and propose Region-Constrained Batch Sampling (RCBS), a simple yet effective batching method that constructs region-homogeneous mini-batches so that users are contrasted primarily against items they could feasibly see. |
Seungho Han; Byeongchang Kim; Jin Yu; |
| 378 | DIGEST: Dynamic Graph Refinement with Dual Contrastive Semantic Transfer for Multimodal Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Multimodal recommendation benefits from leveraging rich content signals such as images and texts to alleviate interaction sparsity, yet existing graph-based approaches are still hindered by (i) noisy user—item edges that are treated as static during training and (ii) inconsistent representation spaces across interaction-driven and modality-induced graph views. To address these issues, we propose DIGEST, a multi-graph framework that propagates trainable ID embeddings on a denoised user—item graph and a fused modality-induced item—item graph, and interleaves message passing with dynamic graph refinement that iteratively reweights existing edges to suppress noisy connections. |
Xiangyu Sai; Meysam Madadi; Sergio Escalera; Yong Xu; |
| 379 | GenRecEdit: Adapting Model Editing for Generative Recommendation with Cold-Start Items Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address these, we propose GenRecEdit, the first model editing framework tailored for generative recommendation. |
Chenglei Shen; Teng Shi; Weijie Yu; Xiao Zhang; Jun Xu; |
| 380 | RES-MR: Risk-Aware Reasoning for Explainable and Safe Medication Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Furthermore, they typically rely on fixed, patient-agnostic safety constraints, neglecting the heterogeneity of risk tolerance across individuals, which poses significant risks to vulnerable populations. To address these gaps, we propose RES-MR, a novel Risk-aware Reasoning framework for Explainable and Safe Medication Recommendation using LLMs. |
Cong Wang; Jin Li; Shoujin Wang; Yishuo Li; Huilin Gu; Wenpeng Lu; |
| 381 | Generative Bid Shading in Real-Time Bidding Advertising Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Additionally, existing discretization models of continuous values ignore the dependence between discrete intervals, reducing the model’s error correction ability, while sample selection bias in bidding scenarios presents further challenges for prediction. To address these issues, this paper introduces Generative Bid Shading (GBS), which comprises two primary components: 1) an end-to-end generative model that utilizes an autoregressive approach to generate shading ratios by stepwise residuals, capturing complex value dependencies without relying on predefined priors; and 2) a reward preference alignment system, which incorporates a channel-aware hierarchical dynamic network (CHNet) as the reward model to extract fine-grained features, along with modules for surplus optimization and exploration utility reward alignment, ultimately optimizing both short-term and long-term surplus using group relative policy optimization (GRPO). |
Yinqiu Huang; Hao Ma; Wenshuai Chen; Zongwei Wang; Shuli Wang; Yongqiang Zhang; Xue Wei; Yinhua Zhu; Haitao Wang; Xingxing Wang; |
| 382 | Hierarchical Cluster-based Open-World Graph Active Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Motivated by the observation that identifying an informative region can be easier than finding the most informative example, we propose a novel hierarchical cluster-based algorithm for open-world GAL. |
Yayong Li; Zhengyi Du; Hong Zhang; Jonathan Wilton; Jinran Wu; Zongli Liu; Nan Ye; |
| 383 | SOMIN: Agentic AI for Automating Professional Visibility and Countering Content Homogenization Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This technological limitation leads to poor content performance and fails to distinguish individual professionals from the noise of AI-generated text. To address this deficit, we introduce SOMIN, an Agentic AI-ready marketing ideation and generation platform designed to bridge the gap between individual professional expertise and effective public communication. |
Aleksandr Farseev; Kirill Lepikhin; Kevin Manuel; Maksim Gorodilov; Zhao Kui; Ilya Makarov; Jaime Francisco Maldonado; Ian Cassidy; Ronan Byrne; |
| 384 | Purifying Multimodal Retrieval: Fragment-Level Evidence Selection for RAG Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Fragment-level Evidence Selection for RAG (FES-RAG), a framework that selects atomic multimodal fragments rather than entire documents as grounding evidence. |
Xihang Wang; Zihan Wang; Chengkai Huang; Cao Liu; Ke Zeng; Quan Z. Sheng; Lina Yao; |
| 385 | MEG-RAG: Quantifying Multi-modal Evidence Grounding for Evidence Selection in RAG Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Building on MEG, we introduce MEG-RAG, a framework that trains a multimodal reranker to align retrieved evidence with the semantic anchors of the ground truth. |
Xihang Wang; Zihan Wang; Chengkai Huang; Quan Z. Sheng; Lina Yao; |
| 386 | SkillForge: Forging Domain-Specific, Self-Evolving Agent Skills in Cloud Technical Support Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce SkillForge, a self-evolving framework that closes an end-to-end creation–evaluation–refinement loop. |
Xingyan Liu; Xiyue Luo; Linyu Li; Ganghong Huang; Jianfeng Liu; Honglin Qiao; |
| 387 | SNBot: Modeling Self–Neighborhood Representation Discrepancy for Social Bot Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose SNBot, a novel social bot detection framework that explicitly models the discrepancy between node self-representations and their neighborhood embeddings. |
Qilong Lin; Jingya Zhou; |
| 388 | Request-Only Optimization for Recommendation Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To utilize the rich user signals in the long user history, these models have been scaled up to unprecedented complexity, up to trillions of floating-point operations (TFLOPs) per example. This scale, coupled with the huge amount of training data, necessitates new storage and training algorithms to efficiently improve the quality of these complex recommendation systems.In this paper, we present a Request-Only Optimizations (ROO) training and modeling paradigm. |
Liang Guo; Wei Li; Lucy Liao; Huihui Cheng; Rui Zhang; Yu Shi; Yueming Wang; Yanzun Huang; Keke Zhai; Pengchao Wang; Timothy Shi; Xuan Cao; Shengzhi Wang; Renqin Cai; Zhaojie Gong; Omkar Vichare; Rui Jian; Leon Gao; Shiyan Deng; Xingyu Liu; Xiong Zhang; Fu Li; Wenlei Xie; Bin Wen; Rui Li; Lu Fang; Xing Liu; Jiaqi Zhai; |
| 389 | MuseKG: An Interactive Knowledge Graph Over Museum Collections Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present MuseKG, an interactive knowledge graph system that organises heterogeneous museum data into a typed graph that links objects, people, organisations, images, image-derived labels, and extracted semantic entities within a coherent schema. |
Jinhao Li; Jianzhong Qi; Soyeon Caren Han; Eun-Jung Holden; |
| 390 | Chunk-Wise Quantization for Graph Collaborative Filtering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper proposes GraphQ, a chunk-wise quantization framework for graph collaborative filtering that supports both the training and post-training phases in a unified perspective. |
Kaixi Hu; Peipei Wang; Kaize Shi; Jingling Yuan; Yu Yang; Guandong Xu; Lin Li; |
| 391 | Structural and Disentangled Adaptation of Large Vision Language Models for Multimodal Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, applying LVLMs to recommendation remains challenging due to ( i ) representation misalignment, where gaps between specific domain data and general pre-training lead to unaligned embedding spaces, and ( ii ) gradient conflicts during fine-tuning, where shared adapters cause interference and a lack of discriminative power. To address this, we propose SDA, a lightweight framework for Structural and Disentangled Adaptation, which integrates two components: Cross-Modal Structural Alignment (CMSA) and Modality-Disentangled Adaptation (MoDA). |
Zhongtao Rao; Peilin Zhou; Dading Chong; Zhiwei Chen; Shoujin Wang; Nan Tang; |
| 392 | ATVG: Agentic System for Factually Grounded Travel Advertisement Video Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose ATVG (Agentic system for factually grounded Travel advertisement Video Generation), an agentic system that generates travel video ads from only a city name. |
Byung Eun Jeon; Xiao Bai; Wen Zhang; Jinchao Li; |
| 393 | TAR: Generative Auto-Bidding and Budget Pacing Via Multi-Scale Trajectory Modeling Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This manifests as sparse reward signals and delayed feedback, forcing agents to learn from locally noisy and incomplete signals. We address this core challenge by introducing the Trajectory Auto-Regressive Model (TAR), a generative framework that aligns planning resolution with feedback dynamics. |
Liang Shi; Longxiang Xu; Zhengju Tang; Yundu Huang; Jian Xu; Zhi Yang; |
| 394 | PeerPrism: Peer Evaluation Expertise Vs Review-writing AI Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In practice, evaluative ideas and surface realization may originate from different sources, creating a spectrum of human-AI collaboration. To address this, we introduce PeerPrism, a large-scale benchmark of 20,690 peer reviews explicitly designed to disentangle idea provenance from text provenance. |
Soroush Sadeghian; Alireza Daghighfarsoodeh; Radin Cheraghi; Sajad Ebrahimi; Negar Arabzadeh; Ebrahim Bagheri; |
| 395 | LearnDCG: End-to-End Joint Optimization of Ranker and Loss in Neural Ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce LearnDCG, a differentiable and learnable approximation of NDCG that unifies ranker and loss optimization within a single end-to-end pipeline. |
Mohammad Hossein Saliminabi; Dimitrios Androutsos; Ebrahim Bagheri; |
| 396 | Peerispect: Claim Verification in Scientific Peer Reviews Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present Peerispect, an interactive system that operationalizes claim-level verification in peer reviews by extracting check-worthy claims from peer reviews, retrieving relevant evidence from the manuscript, and verifying the claims through natural language inference. |
Ali Ghorbanpour; Soroush Sadeghian; Alireza Daghighfarsoodeh; Sajad Ebrahimi; Negar Arabzadeh; Seyed Mohammad Hosseini; Ebrahim Bagheri; |
| 397 | Surface-Form Neural Sparse Retrieval: Robust Fuzzy Matching for Industrial Music Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a robust neural sparse retrieval system designed to maximize exploration efficiency. |
Paul Greyson; Zhichao Geng; Wei Zhang; Yang Yang; |
| 398 | Mitigating Structural Overfitting: A Distribution-Aware Rectification Framework for Missing Feature Imputation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, we find that these approaches suffer from structural overfitting, leading to three progressive challenges: 1) performance degradation on disjoint graphs, 2) loss of semantic diversity due to over-smoothing, and 3) feature distribution shift when generalizing to unseen graph structures (inductive tasks). To address these challenges, we introduce the DART framework. |
Yifan Song; Fenglin Yu; Yihong Luo; Xingjian Tao; Siya Qiu; Kai Han; Jing Tang; |
| 399 | OCP: Orthogonal Constrained Projection for Sparse Scaling in Industrial Commodity Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: A key challenge is that traditional Item-Id vocabularies, when subjected to sparse scaling, suffer from low-frequency information interference, which restricts their expressive power for massive item sets and leads to representation collapse. To address this issue, we propose an Orthogonal Constrained Projection method to optimize embedding representation. |
Chen Sun; Beilin Xu; Boheng Tan; Jiacheng Wang; Yuefeng Sun; Rite Bo; Ying He; Yaqiang Zang; Pinghua Gong; |
| 400 | SilverTorch: A Unified Model-based System to Democratize Large-Scale Recommendation on GPUs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present SilverTorch, a model-based serving system that brings all components into one unified model. |
Bi Xue; Hong Wu; Lei Chen; Chao Yang; Yiming Ma; Fei Ding; Zhen Wang; Liang Wang; Xiaoheng Mao; Ke Huang; Xialu Li; Peng Xia; Rui Jian; Yanli Zhao; Yanzun Huang; Yijie Deng; Harry Tran; Ryan Chang; Min Yu; Eric Dong; Jiazhou Wang; Qianqian Zhang; Keke Zhai; Hongzhang Yin; Pawel Garbacki; Zheng Fang; Yiyi Pan; Min Ni; Yang Liu; |
| 401 | Demand-Calibrated Facet Diversification with Deployable Soft Quotas for E-Commerce Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We proposed a supervision-free, deployable framework for demand-calibrated facet diversification. |
Shenghui Xu; Jiang Yu; |
| 402 | Recursive Short-to-Long Generalization for Multi-hop Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Inspired by the classic recursive algorithm, we propose a novel Recursion-based Short-to-Long Generalization (RSLG) reasoning framework, which recursively decomposes long-hop questions into multiple short-hop questions that can be handled by a short-hop reasoning model. |
Mayi Xu; Ke Sun; Jianhao Chen; Qiankun Pi; Guixin Su; Yunfeng Ning; Yongqi Li; Yuanyuan Zhu; Ming Zhong; Jiawei Jiang; Tieyun Qian; |
| 403 | Logit Inflation in ListMLE: Theoretical Analysis and Mitigation Strategies Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We show that score inflation is a structural consequence of the objective’s overlapping softmax normalizers and shift invariance, which generate persistent gradient contributions even after the correct ordering is achieved. |
Riyaz Ahmad Bhat; Jaydeep Sen; |
| 404 | Mitigating Evidence Suppression: Bi-level Active Evidence Injection for Educational Video Understanding Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While a controlled study shows that performance is sensitive to the strength and amount of injected candidate evidence signals, we find that rigid heuristics are insufficient due to sensitivity to token quality. To address this, we propose Bi-level Active Evidence Injection (BAEI), a decoding-time intervention that keeps the LVLM backbone frozen. |
Cheng Liu; Yiping Wang; Quanlong Guan; Chaobo He; Xingyu Zhu; Liangda Fang; |
| 405 | WISE: A Multimodal Search Engine for Visual Scenes, Audio, Objects, Faces, Speech, and Metadata Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present WISE, an open-source audiovisual search engine which integrates a range of multimodal retrieval capabilities into a single practical tool, accessible to users without machine learning expertise. |
Prasanna Sridhar; Horace Lee; David M. S. Pinto; Andrew Zisserman; Abhishek Dutta; |
| 406 | RE-TRIANGLE: Does TRIANGLE Enable Multimodal Alignment Beyond Cosine Similarity in Retrieval? Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: The TRIANGLE framework addresses this by minimizing the area of modality triplets on a hypersphere to enforce holistic alignment. In this reproducibility study, we verify the robustness of this geometric objective for retrieval tasks. |
Arijit Ghosh; Aritra Bandyopadhyay; Chiranjeev Bindra; Jingfen Qiao; |
| 407 | T-RADAR: Simulating Trademark Examination As An Interactive Retrieval Interface for Conflict Risk Assessment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose agentic simulation as an IR interface and demonstrate it in T-RADAR, a trademark clearance system. |
Yongdeuk Seo; Noah Lee; Hyun-seok Min; Sungchul Choi; |
| 408 | Deja Vu in Plots: Leveraging Cross-Session Evidence with Retrieval-Augmented LLMs for Live Streaming Risk Assessment Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Detecting these risks is challenging because harmful actions often accumulate gradually and recur across seemingly unrelated streams. To address this, we propose CS-VAR (Cross-Session Evidence-Aware Retrieval-Augmented Detector) for live streaming risk assessment. |
Yiran Qiao; Xiang Ao; Jing Chen; Yang Liu; Qiwei Zhong; Qing He; |
| 409 | Improving Ad-hoc Search Effectiveness for Conversational Information Retrieval Via Model Merging Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, this mainstream approach is costly—since it requires model re-training—and exhibits catastrophic forgetting, where the model loses its foundational ad-hoc retrieval performance. In this paper, we fill this gap by introducing model merging as a training-free strategy enabling the design of a single retrieval model that operates across both ad-hoc and conversational settings with no additional fine-tuning. |
Ahmed Rayane Kebir; Jose G. Moreno; Lynda Tamine; |
| 410 | Automated Feature Lifecycle Management in Large-Scale Continual Ranking Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present an end-to-end differentiable feature selection mechanism deployed within a production-scale continual learning framework at Flipkart. |
Bhavani Shankar P S V N; Thejus Vm; Surender Kumar; |
| 411 | From Interference to Stability: Adversarial Reliability Correction for Video Moment Retrieval with Relevance Feedback Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In light of these, we introduce AdversaRial Reliability cOrrection netWork (ARROW) for VMR-RF. |
Hao Liu; Yupeng Hu; Kun Wang; Junchao Wang; Ruping Cao; Yutao Yao; Zilu Cai; |
| 412 | Exploring Test-time Scaling Via Prediction Merging on Large-Scale Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose two ways: One is to explore the heterogeneity of different model architectures. |
Fuyuan Lyu; Zhentai Chen; Jingyan Jiang; Lingjie Li; Xing Tang; Xiuqiang He; Xue Liu; |
| 413 | FEDIN: Frequency-Enhanced Deep Interest Network for Click-Through Rate Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a novel empirical observation: user attention scores exhibit distinct spectral entropy distributions when conditioned on positive versus negative target items. |
Zenan Dai; Jinpeng Wang; Junwei Pan; Dapeng Liu; Lei Xiao; Shu-Tao Xia; |
| 414 | SAER: Scalable Assessment of E-commerce Recommendations Using Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present SAER, a two-stage framework (pointwise filtering and pairwise comparison) that uses Large Language Models as judges to evaluate e-commerce recommendation quality. |
Xiang Liu; Manoj Srivatsav Regulagedda; Sapan Patel; Walter Wong; |
| 415 | NBA-Net: Next-Behavior-Aware Network for Intent Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing methods predominantly exploit historical behaviors to infer user’s current intent, yet overlook the influence of forthcoming user behaviors—since current intent being intrinsically coupled with the behaviors users are poised to perform. Motivated by this insight, we propose the Next-Behavior-Aware Network (NBA-Net), grounded in the accurate generation of the next-behavior and its efficient guidance of the main learning process. |
Haoxin Shen; Peng Ying; Wanjie Tao; Jie Liang; Quan Lu; Ning Jiang; |
| 416 | TReB: A Comprehensive Benchmark for Evaluating Table Reasoning Capabilities of Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: One of these challenges is lacking an effective evaluation benchmark fairly reflecting the performances of LLMs on broad table reasoning abilities. In this paper, we fill in this gap by presenting a comprehensive table reasoning benchmark, TReB. |
Ce Li; Xiaofan Liu; Zhiyan Song; Ce Chi; Boshen Shi; Chen Zhao; Guanguang Chang; Zhendong Wang; Kexin Yang; Xing Wang; Chao Deng; Junlan Feng; |
| 417 | Separating Wheat from Chaff: Fine-Grained Defenses Against Poisoning in Multi-Perspective RAG Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While existing RAG defense methods perform well in closed-domain factoid QA, they face critical challenges in multi-perspective scenarios: excessive contradiction filtering leads to perspective homogenization, and insufficient capability to identify implicit stance biases. To address these limitations, we propose Perspective-Shielded RAG (PS-RAG), a defense framework based on hypergraph modeling. |
Yanxiao Zhao; Longzhu He; Li Sun; Sen Su; |
| 418 | HybridSparse: An End-to-End Hybrid Framework for Efficient Large-Scale Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce HybridSparse, an end-to-end hybrid retrieval framework that strengthens sparse–dense interaction across modeling, training, and serving. |
Haotong Bao; Jianjin Zhang; Weihao Han; Xue Wu; Qi Chen; Dongzhe Jiang; Zhengxin Zeng; Mingzheng Li; Hao Sun; Weiwei Deng; Feng Sun; Qi Zhang; |
| 419 | Adaptive Rich-kernelized Contrastive Learning for Capacity Enhancement in Collaborative Filtering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This inherent bottleneck substantially limits the expressive capacity of such models and reduces their ability to capture complex user-item interaction patterns. To overcome this bottleneck, we propose Adaptive Rich-kernelized Contrastive Learning (ARC), which enhances model expressiveness while maintaining the computational efficiency of single-vector retrieval. |
Jie Yang; Ling Luo; Nestor Cabello; Lars Kulik; |
| 420 | Cohesive Group Discovery in Interaction Graphs Under Explicit Density Constraints Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper presents EDQC, an effective framework for cohesive group discovery under explicit density constraints. |
Yu Zhang; Yilong Luo; Mingyuan Ma; Yao Chen; Enqiang Zhu; Jin Xu; Chanjuan Liu; |
| 421 | A Reproducibility Study of Bundle Editing and Bundle Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper presents the first comprehensive reproducibility study of the complete bundle recommendation pipeline. |
Yiran An; Lin Li; Ming Li; Wenxin Ye; Qing Xie; Jimmy Xiangji Huang; |
| 422 | SQuTR: A Robustness Benchmark for Spoken Query to Text Retrieval Under Acoustic Noise Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing evaluation datasets are often limited to simple queries under constrained noise conditions, making them inadequate for assessing the robustness of spoken query retrieval systems under complex acoustic perturbations. To address this limitation, we present SQuTR, a robustness benchmark for spoken query retrieval that includes a large-scale dataset and a unified evaluation protocol. |
Yuejie Li; Ke Yang; Yueying Hua; Bolin Chen; Jianhao Nie; Yueping He; Caixin Kang; |
| 423 | Load-sensitive Selective Pruning in Dense Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce a load-aware selective embedding-pruning method to be used in dense retrieval systems. |
Maria Diana Calugaru; Federico Siciliano; Francesca Pezzuti; Nicola Tonellotto; Fabrizio Silvestri; |
| 424 | G-CoS: An Interpretable Gain-Cost Framework for User Satisfaction Estimation in Generative Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, empirical analysis of real-world data shows that user satisfaction correlates negatively with interaction signals reflecting interaction cost, and positively with response quality. |
Jia-Ling Shi; Zhijing Wu; Yidong Liang; Xian-Ling Mao; |
| 425 | Simulating The Lateral Reader for News Trustworthiness Reports with An Iterative Multi-Agent RAG System Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present an iterative multi-agent Retrieval-Augmented Generation (RAG) system that operationalizes this workflow for the TREC 2025 DRAGUN Track. |
Dake Zhang; Mark D. Smucker; |
| 426 | SCOPE: Scalable Cross-Task Orthogonal Progressive Experts for Multi-Task Learning in Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing Mixture-of-Experts (MoE) approaches often suffer from knowledge entanglement due to indirect sharing, knowledge redundancy among correlated tasks, and scalability bottlenecks caused by super-linear computational growth as the number of tasks increases. To address this, we propose SCOPE (Scalable Cross-Task Orthogonal Progressive Experts). |
Zixian Yang; Wei Xu; Li Li; Zhaokai Huang; You Li; Jianbin Lin; Wenliang Zhong; Can Ye; |
| 427 | Population-Guided Intent-Aware Query Rewriting for Web Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Abstract: Query rewriting is a core component of web search, yet traditional methods mainly rely on large language model (LLM) prompting, fine-tuning, or personalized rewriting based on … |
Yuanzhao Guo; Shuai Zhang; Anqi Li; Jiaming Zhang; Wei Li; Shihao Liu; Daiting Shi; Yuan Tian; |
| 428 | Adaptive Sparsity Optimization with Learnable Soft Top-K and Per-Term Thresholding for Efficient Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper proposes a scheme for optimizing model sparsity through a synergy of adaptive strategies, including learnable soft top-?? |
Wentai Xie; Parker Carlson; Shanxiu He; Tao Yang; |
| 429 | HE-DeepFM: An FHE Inference System for CTR Prediction with Efficient FM Interactions Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present HE-DeepFM, a FHE inference system for CTR prediction. |
Qiyue Su; Hang Gu; Zhiguang Wang; Teng Wang; Zhendong Zheng; Qianyu Cheng; Lei Gong; Chao Wang; |
| 430 | Hybrid Pooling with LLMs Via Relevance Context Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, standard ICL treats examples as independent instances and fails to explicitly capture the underlying relevance criteria of a topic, restricting its ability to generalise to unseen query-document pairs. To address this limitation, we introduce Relevance Context Learning (RCL), a novel framework that leverages human relevance judgements to explicitly model topic-specific relevance criteria. |
David Otero; Javier Parapar; |
| 431 | Lost in The Evidence? Reproducing Document Position and Context Size Effects in RAG Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, empirical findings remain inconsistent and hard to reproduce across models, datasets, and evaluation protocols. In this paper, we present a systematic reproducibility study that revisits these claims and examines how they evolve with contemporary LLMs under a controlled evaluation framework. |
Jorge Gab\'{\i}n; Anxo P\'{e}rez; Javier Parapar; |
| 432 | Cheaper Is Better: A Discount-Aware Network for Conversion Rate Prediction in E-commerce Recommendation System Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we introduce the Discount-Aware Network (DANet) to model the relationship between item discount rates and CVR. |
Ruocong Tang; Yang Huang; Xing Fang; Chenyi Yan; Chuike Sun; Jing Wang; |
| 433 | NeurPIU: Neurobiologically Inspired Personalized Intent Understanding in Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Based on the observation, we first identify its problem as the ignore of users’ mental states, in which the complicated mental elements, unclear functional rules, and evolving mental states pose obstacles to approach the problem. Therefore, in this paper, we introduce the theory of the mentalizing network in the human brain and propose a neurobiologically inspired framework, i.e., NeurPIU, that endows LLMs with human-like mentalizing capabilities for personalized intent understanding. |
Zenghua Liao; Jinzhi Liao; Xiang Zhao; |
| 434 | Formalized Information Needs Improve Large-Language-Model Relevance Judgments Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We compare assessors using synthetically formalized topics against the LLM-default query-only assessor on the~2019/2020~editions of TREC Deep Learning and Robust04. |
J\{u}ri Keller; Maik Fr\{o}be; Bj\{o}rn Engelmann; Fabian Haak; Timo Breuer; Birger Larsen; Philipp Schaer; |
| 435 | ExDR: Explanation-driven Dynamic Retrieval Enhancement for Multimodal Fake News Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose ExDR—an Explanation-driven Dynamic Retrieval-Augmented Generation framework for Multimodal Fake News Detection. |
Guoxuan Ding; Yuqing Li; Ziyan Zhou; Zheng Lin; Daren Zha; Jiangnan Li; |
| 436 | From Raw Features to Effective Embeddings: A Three-Stage Approach for Multimodal Recipe Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose TESMR, a 3-stage framework for recipe recommendation that progressively refines raw multimodal features into effective embeddings through: (1) content-based enhancement using foundation models with multimodal comprehension, (2) relation-based enhancement via message propagation over user–recipe interactions, and (3) learning-based enhancement through contrastive learning with learnable embeddings. |
Jeeho Shin; Kyungho Kim; Kijung Shin; |
| 437 | Toward A Telugu Question Answering Dataset for Agricultural IR to Serve Marginal Farmers in Drought-Prone Rayalaseema: A Gap Analysis and Roadmap Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We examine the current landscape of agricultural information retrieval resources for Telugu-speaking farmers and find a critical gap: while agricultural QA datasets exist for Hindi, Tamil, and Bengali, no Telugu agricultural QA benchmark exists, let alone one addressing drought-specific contexts. We document available information sources, identify where they fall short, and propose a methodology for constructing a Telugu Agricultural QA dataset using Kisan Call Center records and regional extension materials. |
Piyush Joshi; Priyansh Singhal; |
| 438 | GenFacet: End-to-End Generative Faceted Search Via Multi-Task Preference Alignment in E-Commerce Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we introduce GenFacet, an industrial-grade, end-to-end generative framework deployed at JD.com. |
Zhouwei Zhai; Min Yang; Jin Li; |
| 439 | LUMI: Unsupervised Intent Clustering with Multiple Pseudo-Labels Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose an intuitive, training-free and label-free method for intent clustering in conversational search. |
I-Fan Lin; Faegheh Hasibi; Suzan Verberne; |
| 440 | Behavioral Feature Boosting Via Substitute Relationships for E-commerce Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: On E-commerce platforms, new products often suffer from the cold-start problem: limited interaction data reduces their search visibility and hurts relevance ranking. To address this, we propose a simple yet effective behavior feature boosting method that leverages substitute relationships among products (BFS). |
Chaosheng Dong; Michinari Momma; Yijia Wang; Yan Gao; Yi Sun; |
| 441 | Reproducing Complex Set-Compositional Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We conduct a reproducibility study to benchmark major retrieval families and reasoning-targeted methods on QUEST and QUEST+Variants, and introduce LIMIT+, a controlled benchmark where relevance depends on arbitrary attribute predicates and constraint satisfaction, and less on pretrained knowledge. |
Vincent Degenhart; Dewi Timman; Arjen de Vries; Faegheh Hasibi; Mohanna Hoveyda; |
| 442 | CommonWhy: A Dataset for Evaluating Entity-Based Causal Commonsense Reasoning in Large Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we introduce CommonWhy, a dataset of 15,000 why questions designed to evaluate entity-based commonsense reasoning about causal relationships in LLMs. |
Armin Toroghi; Faeze Moradi Kalarde; Scott Sanner; |
| 443 | StAR: Adaptive Structure-Aware Reranking for Semantic–Structural Alignment in GraphRAG Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce StAR (Structure-Aware Reranking), a plug-and-play reranking module that injects hyperbolic structural similarity into GraphRAG ranking. |
Junghyun Oh; Sungsu Lim; |
| 444 | Calibrate Your Items! Train CTR Models in Two Steps for Robust Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we explore a particularly important and understudied calibration slice: item-level calibration. |
Yi Han; Jay Fleischman; |
| 445 | CLAX: Fast and Flexible Neural Click Models in JAX Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: CLAX is a JAX-based library that implements classic click models using modern gradient-based optimization. |
Philipp Hager; Onno Zoeter; Maarten de Rijke; |
| 446 | Reward Shaping for Robust Refusal in Small Language Models for Retrieval-Augmented Question Answering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In the presence of distractor documents, instruction-tuned models demonstrate inconsistent performance, with answer accuracy metrics deteriorating in most cases. To mitigate this behavior, we introduce Reward Shaping for Refusal and Reasoning (RSRR), a reinforcement learning framework that teaches LMs to reason step-by-step over multiple documents and to refuse to answer when evidence is insufficient. |
Thilina C. Rajapakse; Maarten de Rijke; |
| 447 | HuffmanEmbed: Using Huffman Coding for Embedding Table Compression in Deep Learning Recommendation Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper introduces HuffmanEmbed, a novel embedding table compression framework that improves accuracy at a given compression ratio. |
Chaoyi Jiang; Abdulla Alshabanah; Hossein Entezari Zarch; Keshav Balasubramanian; Murali Annavaram; |
| 448 | GNN-Based Item Indexing for LLM-Enhanced Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Traditional metadata-based identifiers introduce length variability and semantic ambiguity, whereas existing collaborative indexing (CID) approaches often neglect item attributes, show limited cross-dataset generalizability, and incur high computational cost at scale. To address these limitations, we propose a Graph Neural Network (GNN)–based item indexing framework with three coordinated innovations. |
Senlin Mao; Ji Zhang; Peng Zhang; Ze Wang; Xiaoyao Zheng; Jia Wang; |
| 449 | Time-Interval-Aware Disentangled Expert Modeling for Next-Basket Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing NBR methods generally suffer from two limitations: (1) they often entangle these conflicting motives within a single representation, causing habits to overshadow discovery, and (2) they rely on discrete sequential modeling that ignores continuous-time intervals and item-specific periodicities. In this paper, we propose a novel solution named Time-Interval Disentangled Experts (TIDE) to address these challenges. |
Zhiying Deng; Yuan Fu; Usman Farooq; Ziwei Tian; Wei Liu; Jianjun Li; |
| 450 | Constructing Hard-Positive Query–Document Pairs for Dense Retrieval Via Phrase Representativeness Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This suggests a practical route to hard positives: generate relevant queries that rely on document phrases that receive low ranks under these logits. We therefore define model-specific Token and Phrase Representativeness Scores (TRS/PRS) to discover tokens and key phrases that appear in a document but are poorly expressed by its embedding. |
Zhanyu Wu; Richong Zhang; Zhijie Nie; |
| 451 | Beyond Top-e: Simulation-Based Interactive Evaluation for Query Suggestions Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While online experiments provide realistic behavioural signals, they are costly, difficult to scale, and often irreproducible. To bridge this gap, we introduce SIQSE (Simulation-based Interactive Query Suggestion Evaluation), a framework that models query reformulation as an interactive selection task performed by a simulated user. |
Jorge Gab\'{\i}n; Javier Parapar; Xi Wang; |
| 452 | TM-Bench: Benchmarking Large Language Models on Low-Resource Traditional Mongolian Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we introduce TM-Bench, the first comprehensive benchmark for LLMs on Traditional Mongolian. |
Zhenjie Gao; Feilong Bao; Aruukhan Bai; Ruichen Hou; Xieqi Ji; Dabalgan Wang; Hugjil Ming; Yuan Li; |
| 453 | MSW-NTM: A Spherical Wasserstein Autoencoder for Multimodal Neural Topic Modeling with LLM-Guided Topic Refinement Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing multimodal neural topic models, typically based on standard Variational Autoencoders (VAEs), often suffer from posterior collapse and fail to account for the directional nature of L2-normalized embeddings. To address these limitations, We propose MSW-NTM (Multimodal Spherical Wasserstein Neural Topic Model) which is built upon a Wasserstein Autoencoder that performs topic modeling on the unit hypersphere. |
Dayu Guo; Zhiwen Luo; Nizar Bouguila; Wentao Fan; |
| 454 | Attention Grounded Enhancement for Visual Document Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To improve fine-grained relevance modeling, we propose a Attention-Grounded REtriever Enhancement (AGREE) framework. |
Wanqing Cui; Wei Huang; Yazhi Guo; Yibo Hu; Meiguang Jin; Junfeng Ma; Keping Bi; |
| 455 | Verbalizing LightGCN: Direct Learning of Textual Representations from User-Item Interaction Graph Via LLMs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose VerbaLightGCN, a novel LLM-based recommendation framework that integrates the semantic understanding of LLMs with user-item interaction modeling. |
Manh-Khanh Ngo Huu; Hady W. Lauw; |
| 456 | Consensus-Anchored Expansion for Noise-Resilient Re-ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We observe that these failure modes are complementary, with relevant passages concentrated in regions where both retrievers score highly, while single-retriever confidence proves unreliable. Building on this insight, we propose AgreRank, a re-ranking method that leverages sparse-dense consensus for query expansion. |
Nischal Subedi; Cencheng Shen; |
| 457 | Bridging LLM Embeddings and VAE Parameters for Disentangled Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Disentangled recommendation within the Variational Autoencoder (VAE) framework aims to capture multiple user interests. |
Nhu-Thuat Tran; Hady W. Lauw; |
| 458 | Selective Distillation for Continual Named Entity Recognition with Memory Replay Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing methods tend to ignore class imbalance problem that resulted in consistently performance degradation when encounter rare and new class of entity. To address this problem, we propose a unified framework to defy forgetting through selective knowledge distillation by neurons and memory replay. |
Weihua Wang; Yue Cao; Feilong Bao; |
| 459 | ViCSR: A Large-scale Benchmark and Lightweight Two-Stage Framework for Vietnamese Case-to-Statute Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Case-to-Statute Retrieval—identifying applicable statutes from case facts—is essential for judicial efficiency but remains underexplored in low-resource languages like Vietnamese, where annotated legal corpora are scarce and domain-specific challenges persist. To advance research in this setting, we introduce ViCSR, a new benchmark of 10,000 Vietnamese criminal judgments and 1,122 statutory articles with citation-based relevance labels. |
Minh-Hien Nguyen; Khanh Huyen Nguyen; Tan-Minh Nguyen; Hoang-Quynh Le; Thi-Hai-Yen Vuong; |
| 460 | Every Preference Has Its Strength: Injecting Ordinal Semantics Into LLM-Based Recommenders Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Ordinal Semantic Anchoring (OSA), a hybrid CF-LLM framework that explicitly incorporates preference strength by modeling interaction-level user feedback. |
Jiwon Jeong; Donghee Han; Sungrae Hong; Woosung Kang; Mun Yong Yi; |
| 461 | Beyond Dense Connectivity: Explicit Sparsity for Scalable Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose SSR (Explicit Sparsity for Scalable Recommendation), a framework that incorporates sparsity explicitly into the architecture. |
Yantao Yu; Sen Qiao; Lei Shen; Bing Wang; Xiaoyi Zeng; |
| 462 | Localization Boosting for Growth Markets: Mitigating Cross-Locale Behavioral Bias in Learning-to-Rank Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose a multi-objective framework combining behavioral supervision, VLM-derived relevance signals, and locale-aware boosting. |
Suryaa Veerabathiran Seran; Ashwin Naresh Kumar; Tracy Holloway King; Jing Zheng; |
| 463 | FAIR-RAG: An End-to-End Framework for Mitigating Political Bias Through Fair Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To empirically demonstrate this amplification, we first analyze 16,254 documents from the C4 dataset and 24,300 LLM-generated responses, revealing significant left-leaning and supportive stance bias that can propagate strongly from retrieval to generation. To mitigate this amplification of political bias, we propose FAIR-RAG, an end-to-end framework integrating (1) multi-LLM persona-based annotation, (2) a vector database with political-stance metadata, and (3) a multi-stage fairness engine designed for each of the three stages in RAG systems. |
Jaebeom You; Kisung Lee; Hyuk-Yoon Kwon; |
| 464 | SECRET: SEarch Query Classification with Label RETrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Moreover, such classifiers trained to make predictions over a fixed set of labels, cannot be easily extended to incorporate emerging categories. To overcome these challenges, we propose to use contrastive learning for category retrieval with comprehensive label descriptions. |
Anna Tigunova; Ghadir Eraisha; Ahmed Ragab; |
| 465 | Unsupervised 2D Image-Based 3D Model Retrieval Via Decision Boundary Alignment and Graph Semantic Propagation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose a unified framework that integrates Category-Aligned Sampling (CAS), Decision Boundary Alignment (DBA), and Graph Semantic Propagation (GSP) into a single optimization paradigm. |
Nian Hu; Yibo Zhao; Xinhui Li; Chen Li; Cong Liu; Zan Gao; |
| 466 | M-DaQ: Retrieving Samples with Multilingual Diversity and Quality for Instruction Fine-Tuning Datasets Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Multilingual instruction fine-tuning (IFT) empowers large language models to generalize across diverse linguistic and cultural contexts; however, high-quality, systematically curated multilingual IFT datasets remain scarce. To address this gap, we propose M-DaQ (Multilingual Diversity and Quality), a diversity-aware sampling framework that jointly optimizes instruction-response quality and cross-lingual semantic diversity. |
Chunguang Zhao; Yilun Liu; Pufan Zeng; Yuanchang Luo; Shimin Tao; Minggui He; Weibin Meng; Song Xu; Chen Liu; Hongxia Ma; Li Zhang; Boxing Chen; Daimeng Wei; |
| 467 | Beyond Item IDs: Scaling Short-Form-Video Recommendation Via Semantic-Native Long Sequence Modeling Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a production-deployed framework for modeling ultra-long user behavior sequences at a billion-user scale. |
Ruixiao Sun; Diego Uribe Mora; Zhimeng Jiang; Yuanzhen Lin; Jiarui Wang; Yuening Li; Danfeng Guo; Zhizhong Chen; Chuan He; Liang Liu; |
| 468 | An Analysis of Attribute Utilization in Side Information-integrated Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a shuffle-based analysis that examines the functional contribution of item attributes across datasets, models, and fusion strategies by randomly breaking item-attribute alignments and observing the resulting performance changes. |
Minje Kim; Wooseung Kang; Suwon Lee; Gun-Woo Kim; Sang-Min Choi; |
| 469 | ERASE – A Real-World Aligned Benchmark for Unlearning in Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present ERASE, a large-scale benchmark for MU in recommender systems designed to align with real-world usage. |
Pierre Sicco Lubitzsch; Maarten de Rijke; Sebastian Schelter; |
| 470 | Efficient Query Region Expansion and Decomposition Based Spatial Range Query Algorithm Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Building upon ZPI, we propose a novel spatial range query algorithm, ZPI-RQ. |
Lianyin Jia; Rongjin Wang; Yingbin Su; Suprio Ray; Mengjuan Li; Jiaman Ding; |
| 471 | An Asymmetric-difference Based Document Exact Similarity Search Algorithm Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing symmetric-difference based exact similarity search algorithm generates excessively large key bounds, resulting in reduced efficiency. To address this issue, we introduce a novel asymmetric-difference, which decomposes the symmetric-difference into an upper-bound difference and a lower-bound difference, thereby achieving tighter key bounds. |
Lianyin Jia; Yongxue Zhao; Mengjuan Li; Xiuxing Li; Ying Jiang; Jiaman Ding; |
| 472 | SID-Coord: Coordinating Semantic IDs for ID-based Ranking in Short-Video Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose SID-Coord, a lightweight Semantic ID framework that incorporates discrete, trainable semantic IDs (SIDs) directly into ID-based ranking models. |
Guowen Li; Yuepeng Zhang; Shunyu Zhang; Yi Zhang; Xiaoze Jiang; Yi Wang; Jingwei Zhuo; |
| 473 | Joint Optimization of Relevance and Engagement in Multi-Task Ranking for E-Commerce with Efficient LLM Supervision Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Optimizing industrial search ranking models solely for user engagement signals often introduces systematic biases, prioritizing popular or price-anchored items that may not satisfy semantic intent. |
Luming Chen; Jiaqi Xi; Raghav Saboo; Martin Wang; Sudeep Das; |
| 474 | MCP-Focus: Leveraging Function-Oriented Document Enhancement for MCP Server Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, MCP server documents are often unstructured and exhibit ambiguous function semantics, making it difficult to align user requirements with server capabilities during retrieval. To address this issue, we propose MCP-Focus, a function-oriented document enhancement framework that produces retrieval-ready MCP server documentation via a multi-stage agentic pipeline for white-box code analysis and document generation. |
Wenchun Jing; Haiyang Shen; Haoran Wang; Qi Liu; Ningyuan Li; Chaoran Luo; Ning Zhang; Yun Ma; |
| 475 | Adaptive Mix Preference Optimization for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose Adaptive Mix Preference Optimization (AMPO), an adaptive alignment framework that mixes likelihood and preference objectives with self-calibrated, sample-wise confidence adjustment. |
Junbo Qi; Yanyan Zou; Xuanhua Yang; Sulong Xu; Ying Sun; Shengjie Li; |
| 476 | GenRec: A Preference-Oriented Generative Framework for Large-Scale Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, scaling it to large-scale industrial systems introduces three challenges: (i) within a single request, the identical model inputs may produce inconsistent outputs due to the pagination request mechanism; (ii) the prohibitive cost of encoding long user behavior sequences with multi-token item representations based on semantic IDs, and (iii) aligning the generative policy with nuanced user preference signals. We present GenRec, a preference-oriented generative framework deployed on the JD App https://www.jd.com that addresses above challenges within a single decoder-only architecture. |
Yanyan Zou; Junbo Qi; Lunsong Huang; Yu Li; Kewei Xu; Jiahao Gao; Binglei Zhao; Xuanhua Yang; Sulong Xu; Shengjie Li; |
| 477 | Hypergraph Diffusion-Based Sequential Ensemble for CTR Prediction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Unfortunately, most of the existing ensemble learning methods for CTR prediction are designed for cross-sectional data and neglect user historical behavior sequence which is important for predicting user’s future click behavior. To address the above issues, we propose a Hypergraph Diffusion-based Sequential Ensemble framework for CTR prediction (HDSE). |
Zeheng Zhong; Hongzhi Liu; Gong Chen; Boyuan Ren; Guomin Qin; Zhonghai Wu; |
| 478 | LLM-Informed Bayesian Content Exploration in Ultra-Recency Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Rapid content turnover limits the usefulness of historical signals, while sparse early interactions and limited exposure budgets make quality estimation noisy. In this setting, efficient exploration can improve user experience while surfacing emerging creators and trends for long-term platform health.We propose an approach that uses large language models (LLMs) to provide exposure-independent semantic priors for cold-start exploration in such environments. |
Qing Feng; Ivan Ji; Yiting Hua; Shujian Bu; |
| 479 | Item-Targeting with Keyword Enhancement: A Hybrid Approach to Ads Targeting Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While both approaches are effective in isolation, they leave a coverage gap, particularly for broad queries and new items, where item-based targeting often underperforms. This paper presents a novel hybrid targeting framework that bridges this gap by transferring successful query–item associations from keyword targeting into item-based campaigns via semantic embeddings. |
Hua Zou; Dipanwita Saha; Ning Chen; Chen Yang; Xinxin Shu; Abraham Bagherjeiran; |
| 480 | Differentiable Dual Anchor Negative Sampling for Graph-based Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, the single-anchor design introduces noisy negatives, and the discrete hard selection prevents end-to-end optimization. To address these limitations, we propose a differentiable dual-anchor negative sampling framework for graph-based recommendation. |
Xi Wu; Wenzhe Zhang; Kexin Zhao; Jiquan Peng; Jibing Gong; |
| 481 | OxygenREC: An Instruction-Following Generative Framework for E-commerce Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present OxygenREC, an industrial Fast-Slow Thinking recommender: a near-line LLM pipeline produces Contextual Reasoning Instructions, while a high-efficiency encoder-decoder backbone serves fast generation. |
Zhiwei Zhang; Qingyang Li; Ming Zhang; Yanchen Qiao; Zhen Wang; Ziyang Ji; Xiangyu Qian; Shijie Yang; Yanlong Zang; Weijie Ding; Zhi Ma; Zhen Li; Yaqiang Zang; Pinghua Gong; |
| 482 | Inductive Dual-Polarity Modeling Via Static–Dynamic Disentanglement for Dynamic Signed Networks Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose IDP-DSN, an Inductive Dual-Polarity framework for Dynamic Signed Networks. |
Yikang Hou; Junjie Huang; Yijun Ran; Tao Jia; |
| 483 | JARVIS: An Evidence-Grounded Retrieval System for Interpretable Deceptive Reviews Adjudication Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Despite advances in review-level and graph-level detection methods, two pivotal limitations remain: inadequate generalization and lack of interpretability. To address these challenges, we propose JARVIS, a framework providing Judgment via Augmented Retrieval and eVIdence graph Structures. |
Nan Lu; Leyang Li; Yurong Hu; Rui Lin; Shaoyi Xu; |
| 484 | From Time and Place to Preference: LLM-Driven Geo-Temporal Context in Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present a scalable framework that utilizes large language models (LLMs) to create geo-temporal embeddings from timestamps and coarse locations, capturing holidays, seasonal trends, and local/global events. |
Yejin Kim; Shaghayegh Agah; Neeraj Sharma; Mayur Nankani; Maria Peifer; Feifei Peng; H. Howie Huang; Sardar Hamidian; |
| 485 | Evaluation of Agents Under Simulated AI Marketplace Dynamics Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce Marketplace Evaluation, a simulation-based paradigm that evaluates information access systems as participants in a competitive marketplace. |
To Eun Kim; Alireza Salemi; Hamed Zamani; Fernando Diaz; |
| 486 | LTRR: Learning To Rank Retrievers for LLMs Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we explore a query routing approach that dynamically selects from a pool of retrievers based on the query, using both train-free heuristics and learned routing models. |
To Eun Kim; Fernando Diaz; |
| 487 | Preliminary Study of An Evaluation Benchmark for Vision–Language Models in Fashion E-Commerce Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We report an evaluation benchmark for assessing the operational suitability of Vision-Language Models (VLMs) in fashion e-commerce. |
Ryotaro Shimizu; Sai Htaungkham; Shion Sakurai; Yuki Shimizu; |
| 488 | Mind The Metric: Reproducibility and Fair Benchmarking of Spectral Graph Models for Collaborative Filtering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We conduct a reproducibility and replicability study that examines three major families: (i) spectral denoising methods, (ii) graph signal processing (GSP) models, and (iii) spectral propagation approaches. |
Domenico de Gioia; Claudio Pomo; Ludovico Boratto; Tommaso Di Noia; |
| 489 | How Fine-Grained Should A RAG Benchmark Be? A Hierarchical Framework for Synthetic Question Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present HieraRAG, a hierarchical framework for studying granularity in RAG benchmark construction, defining optimal granularity as the level that maximizes discriminative power (the standard deviation of generation quality across categories) within a given RAG configuration. |
Chase M. Fensore; Kaustubh Dhole; Jason Fan; Eugene Agichtein; Joyce C. Ho; |
| 490 | Dual-Diffusional Generative Fashion Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Moreover, these models typically generate only item images, providing limited interpretability. To address these limitations, we propose DualFashion, a Dual-Diffusional Generative Fashion Recommendation Architecture that jointly models image and text modalities for personalized and explainable recommendation. |
Mingzhe Yu; Lei Wu; Qianru Sun; Yunshan Ma; |
| 491 | OpeNTF2: Fairness-aware Graph Neural Team Formation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We contribute OpeNTF2, a unified framework for neural team formation, which achieves state-of-the-art efficacy while improving efficiency in recommending teams of experts who are almost surely successful at completing complex tasks. |
Hamed Loghmani; Md Jamil Ahmed; Kap Thang; Gabriel Rueda; Hossein Fani; |
| 492 | Contrastive Flow Matching for Collaborative Filtering Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: (2) Flow matching is highly sensitive to its input, so input-level perturbations for view construction can change the generated outcome, making positive pairs semantically mismatched. To address these issues, we propose CoFlowCF, a contrastive flow matching framework for collaborative filtering. |
Wangyu Jin; Jiansheng Qian; Wenwen Xia; Hongliang He; Guanfeng Liu; Pengpeng Zhao; |
| 493 | Financial Transaction Retrieval and Contextual Evidence for Knowledge-Grounded Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As general-purpose LLMs struggle with time-distributed tabular data, production stacks still depend on specialized tabular and sequence models that are limited in transferability and require labeled data. To address this, we introduce FinTRACE, a retrieval-first architecture that converts raw transactions into reusable feature representations, applies rule-based detectors, and stores the resulting signals in a behavioral knowledge base with graded associations to the objectives of downstream tasks. |
Artem Sakhno; Daniil Tomilov; Yuliana Shakhvalieva; Inessa Fedorova; Daria Ruzanova; Omar Zoloev; Andrey Savchenko; Maksim Makarenko; |
| 494 | Reasoning-Grounded Intent Injection for Generative Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Industrial generative recommendation systems operating over discrete Semantic IDs (SIDs) are largely behavior-driven, and thus struggle to proactively activate latent demand before explicit user signals emerge, leading to intent cold-start. To address this, we propose RIGER (Reasoning-grounded Intent injection for GE nerative Recommendation), a deployable two-stage framework that integrates offline large language model (LLM) reasoning into an online generative recommender under strict latency constraints. |
Xusong Chen; Peini Guo; Fang Liu; Yu Zhao; Yiyang Hu; Mengqin Que; Peng Li; Haoran Wang; Zhiwei Fang; Wenlong Chen; Changping Peng; Ching Law; |
| 495 | MathMex-PDF: Towards Accessible Visual Mathematics Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Retrieval uses a late-fusion pipeline that combines structure-aware formula matching with dense text embeddings and applies reciprocal rank fusion to adopt modality-specific rankings. |
Nathaniel Serrano; Lucas Matheson; Clayton Durepos; Behrooz Mansouri; |
| 496 | LLM-Click Agreement: Harmonizing Implicit Feedback and Semantic Judgments for Enterprise Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present LLM-Click Agreement Labeling, an industrial-scale supervision strategy that retains only those query–message pairs where click-based labels and LLM-generated labels agree. |
Avinash Kumar; Hardi Rathod; Rohan Mallick; Swarnangsu Acharyya; |
| 497 | Efficient Multivector Retrieval with Token-Aware Clustering and Hierarchical Indexing Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we introduce Tachiom, a multivector retrieval system that exploits token-level structure to significantly accelerate both clustering and retrieval. |
Silvio Martinico; Franco Maria Nardini; Cosimo Rulli; Rossano Venturini; |
| 498 | GeoSearch: Augmenting Worldwide Geolocalization with Web-Scale Reverse Image Search and Image Matching Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose GeoSearch, an open-world geolocation framework that integrates web-scale reverse image search into the RAG pipeline. |
Tung-Duong Le-Duc; Hoang-Quoc Nguyen-Son; Minh-Son Dao; |
| 499 | Beyond Maintenance: A Benchmark and Multi-Agent Framework for Repository-Usage Code Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Building on RUCCE, we propose RUCACoder, a closed-loop multi-agent framework with a Retriever for hierarchical repository exploration, a Verifier for reranking and validation, and a Coder for feedback-driven script synthesis. |
Kaitao Lin; Songwen Gong; Adam Jatowt; Jiexin Wang; Yi Cai; |
| 500 | CSMAD: Hallucination Detection Via Multi-Agent Debate with NLI-Verified Contradictory Statements Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing approaches either collaborate which amplifies shared overconfidence, or adopt adversarial preset stances, that can inject incorrect information complicating decision making. To address this, we propose Contradictory Statement Multi-Agent Debate (CSMAD), a multi-agent framework that creates structured disagreement by generating a contradictory claim for each input claim. |
Swapnil Gupta; Akshay Verma; Khushi Gupta; Prateek Sircar; Deepak Gupta; |
| 501 | COMET: Compatibility-Oriented Multi-modal Embedding Transformer for Visual Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing methods often struggle with heterogeneous compatibility rules (e.g., sofa-table vs. sofa-curtain) and an over-reliance on global visual features, missing critical textual cues like style or material. To address these limitations, we introduce COMET (Compatibility-Oriented Multi-modal Embedding Transformer), a scalable, vision-language framework for visual recommendations. |
Dween Rabius Sanny; Prateek Sircar; Deepak Gupta; |
| 502 | Selective Constraint Learning for Unsupervised Cross-Domain Image Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, such internally induced supervision tends to impose an upper bound on achievable retrieval performance, as it lacks stable semantic references to support reliable category-level correspondence across domains. To address these limitations, we propose Selective Constraint Learning (SCL), a framework that introduces external semantic guidance as a stable prior for unsupervised cross-domain image retrieval. |
Wensi Fang; Xiaodan Zhang; Xiaoyu Lian; Qiang Li; Shuai L\{u}; |
| 503 | Following The Eye-Tracking Evidence: Established Web-Search Assumptions Fail in Carousel Interfaces Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Our findings show that many user behavior assumptions, especially concerning examination patterns, do not transfer from web search interfaces to carousel recommendation settings. Our work shows that the field lacks a reliable foundation on which to build models of user behavior with these interfaces. |
Jingwei Kang; Maarten de Rijke; Harrie Oosterhuis; |
| 504 | A Voronoi Cell Formulation for Principled Token Pruning in Late-Interaction Retrieval Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Past works address this by attempting to prune low-importance token embeddings based on statistical and empirical measures, but they often either lack formal grounding or are ineffective. To address these shortcomings, we introduce a framework grounded in hyperspace geometry and cast token pruning as a Voronoi cell estimation problem in the embedding space. |
Yash Kankanampati; Yuxuan Zong; Nadi Tomeh; Benjamin Piwowarski; Joseph Le Roux; |
| 505 | Graph Diffusion Gated Embeddings for Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, collisions and hard bucket assignments can degrade both performance and personalization. We introduce Graph Diffusion Gated Embeddings (GDE), which precompute multi-scale, type-separated heat-kernel diffusions from a small set of user/item seeds (via HK-relax) and convert them into degree-corrected probabilistic gating distributions. |
Seungcheol Lee; Taeyoung Roh; Jiho Seo; Soohyun Lim; |
| 506 | Tokens to Types: Context Editing with Selective Entity Abstraction for Grounded Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While modern reasoning models improve general response quality, they fail to resolve these underlying prior knowledge biases even when generating a high volume of costly thinking tokens. We propose a context-editing framework that addresses this by performing selective abstraction over entities that appear in both the context and the question. |
Rounak Sharma; Debabrata Mahapatra; Shiv Kumar Saini; |
| 507 | Debiased Multimodal Personality Understanding Through Dual Causal Intervention Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Learning such spurious associations between multimodal features and traits may lead to unfair personality understanding. In this work, we construct a Structural Causal Model (SCM) to analyze the impact of these biases from a causal perspective, and propose a novel Dual Causal Adjustment Network (DCAN) to mitigate the interference of subject attributes on personality understanding. |
Yangfu Zhu; Zitong Han; Nianwen Ning; Yuting Wei; Yuandong Wang; Hang Feng; Zhenzhou Shao; |
| 508 | From Tokens to Concepts: Leveraging SAE for SPLADE Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, they rely on the underlying backbone vocabulary, which might hinder performance (polysemicity and synonymy) and pose a challenge for multi-lingual and multi-modal usages. To solve this limitation, we propose to replace the backbone vocabulary with a latent space of semantic concepts learned using Sparse Auto-Encoders (SAE). |
Yuxuan Zong; Mathias Vast; Basile Van Cooten; Laure Soulier; Benjamin Piwowarski; |
| 509 | Evaluation Validity in Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We discuss practical ways to discuss, measure, and improve the validity of evaluations in a range of settings. |
Paul Thomas; Nick Craswell; Mark Sanderson; Seth Spielman; Robert Sim; Ryen W White; |
| 510 | ExODRec: An Explainable Framework for Outlier Detection Model Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: The unsupervised outlier model recommendation (UOMR) problem remains underexplored and presents two major challenges: severe cold-start constraints and the lack of explainable model recommendation. To address these issues, we propose ExODRec, an explainable two-stage meta-learning framework for UOMR. |
Saba Fathi Rabooki; Ziqi Xu; Elham Naghizade; |
| 511 | MuPP: Multi-Preference Padding for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, the former approach demands high computational resources due to redundant generation of user representations, while the latter reinforces redundant item embeddings, which limits the expressiveness of user preferences and reduces representational diversity by repeatedly inserting identical items. To overcome these limitations, we propose a novel padding method called Multi-Preference Padding (MuPP). |
Wooseung Kang; Minje Kim; Suwon Lee; Gun-Woo Kim; Sang-Min Choi; |
| 512 | Auditing Query Drift: Do Users Actually Benefit from Pseudo-Relevance Feedback? Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Current Selective PRF (sPRF) strategies attempt to mitigate this by predicting when PRF will help, but they rely on offline metrics that may not reflect actual user preferences. To bridge this gap, we propose a participatory auditing strategy that evaluates PRF’s real-world impact through natural user interactions. |
Zeyan Liang; Graham McDonald; Iadh Ounis; |
| 513 | When RAG Disagrees: Detecting Latent Epistemic Conflict Via Logit Interactions Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Current approaches to predicting these conflicts rely on computationally expensive supervised probing or black-box adjudication, introducing significant latency. We propose a training-free mechanistic predictor, the Interaction Score, derived from the parametric confidence of the model and its semantic alignment with the retrieved context. |
Saisab Sadhu; Dwaipayan Roy; Tanmay Basu; |
| 514 | Same Image, Different Meanings: Toward Retrieval of Context-Dependent Meanings Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: A scene of two people in the rain can convey hope and warmth in a reunion story or sorrow and finality in a farewell story. We investigate this context-dependent nature of image meaning and its implications for retrieval. |
Ayuto Tsutsumi; Ryosuke Kohita; |
| 515 | Exploring and Improving Cross- and Multi-Domain Personalized Question Answering Via A Dual Retrieval Augmentation Approach Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we use the LaMP-QA benchmark, where user profiles span diverse Stack Exchange domains (Arts & Entertainment, Lifestyle & Personal Development, and Society & Culture), to systematically investigate how domain composition of the user profile affects personalization performance. |
Ozel Yilmazel; Hamed Zamani; James Allan; |
| 516 | MMRM: A Multiplex Multimodal Representation Model for Product Ranking in E-commerce Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Despite their efficacy, these methods face two primary limitations: (1) they rely on a single collaborative signal for MLLM fine-tuning, failing to exploit the heterogeneous signals essential for multitask ranking; and (2) they treat multimodal representations as regular item features in ranking models, underutilizing their latent potential for user behavior modeling. To address these challenges, we propose the Multiplex Multimodal Representation Model (MMRM), a unified framework that aligns MLLMs with diverse collaborative signals. |
Zhen-Lin Chen; Maosen Sheng; Peng Lin; Jianmin Chen; Zhuojian Xiao; Dongyue Wang; Xiwei Zhao; |
| 517 | Gated Bidirectional Linear Attention for Generative Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Gated Bidirectional Linear Attention (GBLA), a linear-time bidirectional attention layer that extends kernelized linear attention with three lightweight components: local causal mixing (Conv1D), sequence-level key gating for soft forgetting, and a gated RMSNorm output. |
Artem Matveev; Vladislav Tytskiy; Sergei Makeev; Sergei Liamaev; |
| 518 | Unifying User Satisfaction and Creator Incentive: A Constrained Optimization Framework for Short-Video Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Second, optimizing creator incentive requires globally coordinated decisions across requests, making real-time serving infeasible. To address these challenges, we formulate the joint maximization as a constrained optimization problem that unifies the two heterogeneous objectives. |
Xiaoru Qu; Dingyi Zhang; Zhangxi Yan; Peng Zhang; Hu Liu; Yang Zou; Jian Liang; Kaiqiao Zhan; |
| 519 | As It Was: Aligning LLM Search Evaluation with Historical User Preferences Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce a behavior-grounded LLM judge that augments each SERP item with a lightweight, auditable behavioral prior in the form of a Query–Relevance–Impressions (QRI) card. |
Ali Vardasbi; Gustavo Penha; Enrico Palumbo; Claudia Hauff; Hugues Bouchard; Mounia Lalmas; |
| 520 | Labadain Chat: A Conversational Agent for The Tetun Language Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This study presents Labadain Chat, a conversational agent for Tetun, a low-resource language spoken by over 932,000 people in Timor-Leste. |
Gabriel de Jesus; S\'{e}rgio Nunes; |
| 521 | Retrieval for User-Centered Translation: Lessons from RAG-based Tools for Low-Resource Domains Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present lessons learned from Tulun, a retrieval-augmented system combining neural MT with LLM post-editing, guided by user-configurable translation memories and glossaries. |
Raphael Merx; Ekaterina Vylomova; |
| 522 | An Eye Tracking Study: Are AI Overviews Changing Search Behavior? Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We conduct a lab eye-tracking study to examine how users interact with search engines that place Generative Artificial Intelligence (GenAI) results above traditionally ranked search results, also known as ten blue links, and we use an engagement scale to evaluate their experience. Our aim is to study how users interact with search engine interfaces that incorporate GenAI content, assess users’ willingness to scroll past GenAI content to view the traditional search results, and explore how these interactions differ from existing literature on scanning search engine result pages. |
Sara Allawati; Dana McKay; Mark Sanderson; Paul Thomas; Johanne Trippas; |
| 523 | GCA-KBQA: A Step-Wise Logical Form Generation Approach for KBQA with Knowledge-Assisted Calibration Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, generating complete LFs with LLMs continues to pose a challenge due to the complexity of the required graph structures and constraints, leading to the significant issue of non-executability. To address these challenges, we propose GCA-KBQA, a step-wise fine-tuned LLM-based framework that employs hop-wise generation, knowledge-assisted calibration, and path-level assembly to construct complete LFs for KBQA. |
Ranran Bu; Jian Cao; Jianqi Gao; Jinghua Tang; Shiyou Qian; Hongming Cai; |
| 524 | Diagnosing Identifiability in Two-Tower Models for Unbiased Learning to Rank Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a practical identifiability diagnostic for two-tower models. |
Stan Fris; Philipp Hager; |
| 525 | NumColBERT: Non-Intrusive Numeracy Injection for Late-Interaction Retrieval Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: These approaches intrude upon late-interaction retrieval models such as ColBERT and incur considerable challenges in deployment, latency, and maintainability. To overcome these limitations, we propose NumColBERT, an inference-time non-intrusive method that enhances numerically conditioned retrieval while preserving the original late-interaction mechanism and providing unified scoring across textual and numerical content. |
Haruki Fujimaki; Makoto P. Kato; |
| 526 | MUDY: Multi-Granular Dynamic Candidate Contextualization for Unsupervised Keyphrase Extraction Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose a novel context-centric framework, MUDY, that effectively captures multi-granular contextual salience of candidate keyphrases. |
Hyeongu Kang; Susik Yoon; |
| 527 | Set-Based Cross-Domain Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In contrast, we propose SetCDR, which constructs more effective user representations in the target domain by directly incorporating each user’s source and target history. |
Kyunglim Kim; James Russell Geraci; |
| 528 | Countering Interest Over-Smoothing: Distilling Latent Factors Via Diffusion for Multi-Interest Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This aggregation dilutes the intensity of significant patterns that appear only locally, blending them into a blurry average. To address this, we propose DMI, a model-agnostic diffusion framework that distills precise interests by amplifying co-occurring latent factors across behaviors. |
Yankun Le; Fu Zhang; Haoran Li; Baoyuan Ou; Yingjie Qin; Zhixuan Yang; Ruilong Su; |
| 529 | A Comparative Study of Users’ Information-Seeking Practices Across Search Engines and Generative AI Chatbots Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: With search engines, they follow a ‘Browse and Verify’ approach, prioritizing reliability and source verifiability. With GenAI chatbots, they adopt a ‘Delegate and Consume’ approach, valuing quick summarization and personalized content. |
Elsa Lichtenegger; Aleksandra Urman; Aniko Hannak; |
| 530 | Cross-Domain Interest Representation Learning for Scenario- and Task-Aware Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In our industrial setting, we observe three key phenomena that existing methods rarely address: (i) users’ Cross-Domain Interests (CDI) are weakly exploited, since behavior sequences are often pooled for efficiency, losing transferable cross-domain dependencies; (ii) multi-domain, scenario, and task variations are under-modeled, making it difficult to capture fine-grained complementarities; and (iii) multimodal features remain misaligned with ID features, especially when different domains emphasize different modalities. |
Bokai Lin; Naijun Gao; Yucen Gao; Heng Chang; Cheng Hu; Zhinan Zhang; Xiaofeng Gao; |
| 531 | How Variability Influences Podcast Search: Queries, Transcriptions, and Judges Related Papers Related Patents Related Grants Related Venues Related Experts View Save Abstract: Podcasts have continued to grow in popularity over the last two decades, with more than 4.52 million podcasts and 584 million listeners across the globe in 2025. Developing … |
Watheq Mansour; J. Shane Culpepper; Andrew Yates; Joel Mackenzie; |
| 532 | DocQAC: Adaptive Trie-Guided Decoding for Effective In-Document Query Auto-Completion Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Unlike traditional WebQAC systems, DocQAC can leverage rich document context, having access not only to the partially typed user query and global historical queries, but also the content of the current document itself, and crucially, the document-specific history of user query interactions. To address this setting, we propose a novel adaptive trie-guided decoding framework that uses user query prefixes to softly steer language models toward high-quality completions. |
Rahul Mehta; Kavin R V; Indrajit Pal; Tushar Abhishek; Pawan Goyal; Manish Gupta; |
| 533 | Recasting Web-Scale Query Suggestion As Dense Retrieval: Efficient, Up-to-Date, and Context-Aware Suggestions Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this study, we recast QS as dense retrieval: given a search session, the next query is retrieved from a large index of historical queries using efficient approximate nearest neighbor search. |
Sosuke Nishikawa; Naoki Yoshinaga; Nobuhiro Kaji; |
| 534 | SciCheck: Reasoning Distillation for Biomedical Claim Verification Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, these approaches fall short when applied to real-world verification tasks, where relevant evidence is hard to obtain and efficiency is crucial. To address this, we introduce SciCheck, a novel solution that integrates web evidence retrieval with a process of reasoning distillation. |
Gabriel Pereira; Luciano Barbosa; |
| 535 | An Epistemic Position-Based Click Model: From Interactions to Epistemic Distributions of Relevance and Bias Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: These distributions capture epistemic uncertainty about click probabilities and the underlying effects of attraction and position bias. The main challenge of our approach is its optimization for which we propose approximation and conditioning techniques to provide numerical stability and variance reduction. |
Oscar Rolando Ramirez Milian; Harrie Oosterhuis; |
| 536 | Hierarchical Embedding Fusion for Retrieval-Augmented Code Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present Hierarchical Embedding Fusion (HEF), a two-stage repository representation for code completion: (i) an offline cache that compresses repository chunks into a reusable hierarchy of dense vectors using a small fuser model, and (ii) an online interface that maps a small number of retrieved vectors into learned pseudo-tokens consumed by a code generator. |
Nikita Sorokin; Ivan Sedykh; Valentin Malykh; |
| 537 | Beyond Exposure Diversity: Debiasing News Consumption With Topic-Locality Calibration and Personalized Preview Nudges Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this study, we applied the personalized diversity nudge framework with the goal of expanding readers’ reading coverage in terms of news locality (i.e., domestic and world news). |
Ruixuan Sun; Matthew Zent; Minzhu Zhao; Xinyi Li; Thanmayee Boyapati; Joseph A. Konstan; |
| 538 | RecPFN: Prior-Fitted Networks for In-Context-Based Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce RecPFN, a prior-fitted network that brings in-context learning to sequential recommendation. |
En Zhi Tan; Jia Xiang Lim; Bryan Lijie Chew; Tze Minh Ng; Benjamin Yan Han Yap; |
| 539 | Cognitive Style Shapes Search Behaviours: An FNIRS Study of Exploratory Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Cognitive style, a user’s habitual approach to information processing, offers a promising approach to personalising information searching (IS) systems, yet the underlying style-related search behaviours remain poorly understood. |
Huimin Tang; Boon Giin Lee; Dave Towey; Max L. Wilson; Matthew Pike; |
| 540 | Fusion and Alignment Enhancement with Large Language Models for Tail-item Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose a Fusion and Alignment Enhancement framework with LLMs for Tail-item Sequential Recommendation (FAERec), which improves item representations by generating coherently-fused and structurally consistent embeddings. |
Zhifu Wei; Yizhou Dang; Guibing Guo; Chuang Zhao; Zhu Sun; |
| 541 | From Existence to Exhaustiveness: Unveiling The Compounding Failures of LLMs in Multi-answer Event Temporal Reasoning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In real-world scenarios, entities often simultaneously play multiple roles or exist in multiple states within the same timeframe. To bridge this gap, we introduce MulTR, a comprehensive benchmark designed for multi-answer temporal reasoning from long unstructured contexts. |
Shaojuan Wu; |
| 542 | Coupling Global Context with Kinematic Evolution for Sequential Recommendation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: While existing models have made significant strides in capturing sequential dependencies, they predominantly encode historical behaviors via discrete aggregation, often failing to explicitly model the continuous evolution trends and intrinsic momentum underlying preference shifts. To address this limitation, we propose Kinematic Evolution for Sequential Recommendation (KERec), which models user preference evolution from a continuous kinematic perspective. |
Yueting Yang; Yao Li; Rongmei Zhao; Shenggen Ju; |
| 543 | NuggetIndex: Governed Atomic Retrieval for Maintainable RAG Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose NuggetIndex, a retrieval system that stores atomic information units as managed records, so called nuggets. |
Saber Zerhoudi; Michael Granitzer; Jelena Mitrovi\'{c}; |
| 544 | How Far Are We from Automatically Identifying Violations of The Data Minimization Principle in Privacy Policies? Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Meanwhile, privacy policies are semantically complex and unstructured, hindering accurate extraction of fine-grained data practices and large-scale automated evaluation. To address these issues, we propose DataMini, a human–LLM collaborative evaluation framework for identifying violations of the data minimization principle in privacy policies. |
Ziyan Zhou; Yanru He; Yunchuan Guo; Haoyang Yu; Liang Fang; Fenghua Li; |
| 545 | Stop Using The Wilcoxon Test: Myth, Misconception and Misuse in IR Research Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We argue that this narrative is misleading and harmful. A careful review of Statistics textbooks reveals inconsistencies and omissions in how the assumptions underlying these tests are presented, fostering confusion that has propagated into IR research. |
Juli\'{a}n Urbano; |
| 546 | The Relevance of (Relevant) Entity Presence Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: A deep dive into the published artifacts, plus a comparative study of similar work by the same authors show inconsistencies between the descriptions in the paper and their implementation in the published artifacts, and a leakage of relevance assessments due to oversampling entities that are known to appear in relevant documents only. |
Norbert Boudens; Chris Kamphuis; Arjen P. de Vries; |
| 547 | Reproduction Beyond Benchmarks: ConstBERT and ColBERT-v2 Across Backends and Query Distributions Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We evaluate ColBERT-v2 and ConstBERT across five dimensions, finding that while ConstBERT reproduces within 0.05\% MRR@10 on MS-MARCO, both models show a drop of 86–97\% on long, narrative queries (TREC ToT 2025). |
Utshab Kumar Ghosh; Ashish David; Shubham Chatterjee; |
| 548 | Insights Into The Efficiency of Open-Source Score-at-a-Time Search Engines: A Reproducibility Study Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We identify differences in postings lists due to indexing, ranking functions, and quantization. Thus, we introduce ciffTools for quantizing the ciff indexes used by both search engines; eliminating these differences. |
Katelyn Harlan; Andrew Trotman; Veronica Liesaputra; |
| 549 | A Reproducibility Study of Metacognitive Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address this, Zhou et al. [51] introduced Metacognitive Retrieval Augmented Generation (MetaRAG), a framework inspired by metacognition that enables Large Language Models to critique and refine their reasoning. In this reproducibility paper, we reproduce MetaRAG following its original experimental setup and extend it in two directions: (i) by evaluating the effect of PointWise and ListWise rerankers, and (ii) by comparing with SIM-RAG, which employs a lightweight critic model to stop retrieval. |
Gabriel Iturra Bocaz; Petra Galu\v{s}\v{c}\'{a}kov\'{a}; |
| 550 | Price-Aware Recommender Systems: A Cross-Paradigm Reproducibility Study Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We successfully conducted the first cross-evaluation across six public datasets using standardized metrics (HR@20, MRR@20, NDCG@20). |
Fernando Medina-Quispe; David Contreras Aguilar; Ludovico Boratto; Maria Salam\'{o}; |
| 551 | From Top-1 to Top-K: A Reproducibility Study and Benchmarking of Counterfactual Explanations for Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we systematically reproduce, re-implement, and re-evaluate eleven state-of-the-art CE methods for recommender systems, covering both native explainers (e.g., LIME-RS, SHAP, PRINCE, ACCENT, LXR, GREASE) and specific graph-based explainers originally proposed for GNNs. |
Quang-Huy Nguyen; Thanh-Hai Nguyen; Khac-Manh Thai; Duc-Hoang Pham; Huy-Son Nguyen; Cam-Van Thi Nguyen; Masoud Mansoury; Duc-Trong Le; Hoang-Quynh Le; |
| 552 | KCC: Korean Civil Case Dataset for Legal Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce KCC, a publicly available and reusable benchmark resource for legal IR in Korean civil law. |
Minhan Cho; Soyoung Park; S. Shyam Sundar; Daejin Choi; Jinyoung Han; |
| 553 | Rerankers: A Lightweight Unified Toolkit for Reranking Approaches Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper presents rerankers, a Python library which provides an easy-to-use interface to the most commonly used re-ranking approaches. |
Benjamin Clavi\'{e}; |
| 554 | WebFAQ 2.0: A Multilingual QA Dataset with Mined Hard Negatives for Dense Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce WebFAQ 2.0, a new version of the WebFAQ dataset, containing 198 million FAQ-based natural question-answer pairs across 108 languages. |
Michael Dinzinger; Laura Caspari; Ali Salman; Irvin Topi; Jelena Mitrovi\'{c}; Michael Granitzer; |
| 555 | GONets: A First-Look Into GitHub Organisation Networks Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present an automated data collection pipeline that retrieves organisational and membership information via GitHub’s API, aggregates user-repository contribution data, and constructs large-scale bipartite networks suitable for network analysis and modelling. |
Hridoy Sankar Dutta; Biswadeep Khan; Parth Mitesh Shah; Amit A. Nanavati; |
| 556 | The Alignment Gap: A Benchmark Demonstrating The Lack of Cross-Lingual Mapping in Dialect-Specialized Language Models – The Case of Ehugbo Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Through diagnostic analysis (t-SNE visualizations, tokenization studies, error patterns), we show that regional models lack cross-lingual alignment bridges despite deep language understanding, while global models achieve language invariance through explicit translation supervision. |
Ukachi Agnes Eze-Mbey; Victor Tolulope Olufemi; Athanase Biluge Bahizire; Mikel K. Ngueajio; Prasenjit Mitra; |
| 557 | A Dataset of Cultural Heritage Manipulation on English Wikipedia in The Russo–Ukrainian Context Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present a new dataset of English Wikipedia revision pairs designed to support the systematic study of cultural heritage manipulation in the context of the Russo–Ukrainian conflict. |
Maxime Garambois; Hamest Tamrazyan; Emanuela Boros; |
| 558 | RRE-GTC: A Geo-Temporal Cluster Dataset for Real-Estate Price Estimation and Market-Aware Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, research into Geo-spatial Information Retrieval (GIR) and location-based recommendations is currently impeded by a lack of open, high-quality datasets that capture both the temporal evolution of markets and granular accessibility signals. To bridge this gap, we introduce RRE-GTC (Ru-Real-Estate Geo-Temporal Clusters), an open resource designed to benchmark ranking, pricing, and recommendation algorithms in dynamic spatial contexts. |
Irina Govorova; Aleksandr Alekseitsev; Irina Podlipnova; Meruza Kubentayeva; Yuriy Dorn; |
| 559 | Multilingual and Domain-Agnostic Tip-of-the-Tongue Query Generation for Simulated Evaluation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we construct multilingual ToT test collections for Chinese, Japanese, Korean, and English, using an LLM-based query simulation framework. |
Xuhong He; To Eun Kim; Maik Fr\{o}be; Jaime Arguello; Bhaskar Mitra; Fernando Diaz; |
| 560 | ITIMO: An LLM-empowered Synthesis Dataset for Travel Itinerary Modification Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, existing research mainly focuses on fixed itinerary planning, leaving modification underexplored due to the scarcity of shape need-to-modify itinerary data. To bridge this gap, we formally define the itinerary modification task and propose a general pipeline to construct the corresponding dataset, namely iTIMO. |
Zhuoxuan Huang; Yunshan Ma; Hongyu Zhang; Hua Ma; Zhu Sun; |
| 561 | A Taxonomy and Catalog of SERP Elements Across Web Search Engines Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, analyses of SERP elements often suffer from inconsistent terminology, making cross-study comparisons difficult. To address this issue, this work proposes a comprehensive catalog of SERP elements that standardizes their classification. |
Adelaide Miranda Santos; Carla Teixeira Lopes; |
| 562 | NERBench-Chhattisgarh: A Multi-Family NER Dataset for Low-Resource Indic Languages Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present NERBench-Chhattisgarh, a gold-standard Named Entity Recognition (NER) dataset covering seven under-resourced languages spoken in Central India: Baigani, Chhattisgarhi, Surgujia, Sadri, Kudukh, Halbi, and Gondi. |
Rajesh Kumar Mundotiya; |
| 563 | CTCL: A Cross-Language Benchmark for Matching Patients to Clinical Trials Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We benchmark the cross-lingual retrieval task using 14 large language (embedding) models. |
Maciej Rybinski; Wojciech Kusa; Necva B\{o}l\{u}c\{u}; Georgios Peikos; Aditya Joshi; Sarvnaz Karimi; Aitziber Atutxa; Javier Del Ser; Ahmet B\{o}l\{u}c\{u}; Monica Chierichetti; Pritam Dasgupta; Nicol\'{a}s Jim\'{e}nez Garc\'{\i}a; Borja Pedruzo; Angelika Roma\'{n}ska; Ioulia Symeonidou; |
| 564 | JFinTEB: Japanese Financial Text Embedding Benchmark Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce JFinTEB, the first comprehensive benchmark specifically designed for evaluating Japanese financial text embeddings. |
Masahiro Suzuki; Hiroki Sakaji; |
| 565 | SplitLight: An Exploratory Toolkit for Recommender Systems Datasets and Splits Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Seemingly minor decisions in filtering, handling repeats, cold-start treatment, and splitting strategy design can substantially reorder model rankings and undermine reproducibility and cross-paper comparability. In this paper, we introduce SplitLight, an open-source exploratory toolkit that enables researchers and practitioners designing preprocessing and splitting pipelines or reviewing external artifacts to make these decisions measurable, comparable, and reportable. |
Anna Volodkevich; Dmitry Anikin; Danil Gusak; Anton Klenitskiy; Evgeny Frolov; Alexey Vasilev; |
| 566 | AgentSim: A Platform for Verifiable Agent-Trace Simulation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce AgentSim, an open-source platform for simulating RAG agents. |
Saber Zerhoudi; Michael Granitzer; Jelena Mitrovic; |
| 567 | IIRSim Studio: A Dashboard for User Simulation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: The bottleneck is not the simulation engines themselves, but the lack of infrastructure connecting experiment design, execution, and sharing into a single verifiable workflow. This paper introduces IIRSim Studio, a web-based workbench that addresses this gap through four contributions: (1) a visual environment for composing simulation pipelines on top of simulation frameworks, serving both novices learning simulation concepts and experts piloting large-scale experiments; (2) a component lifecycle that supports authoring, versioning, and sharing custom simulation components through Git-backed storage and runtime injection; (3) a provenance model based on experiment bundles and environment templates that makes the scope of replication explicit; and (4) a shared-task workflow, demonstrated through the re-deployment of a Sim4IA micro-task. |
Saber Zerhoudi; Adam Roegiest; Michael Granitzer; |
| 568 | SimEval-IR: A Unified Toolkit and Benchmark Suite for Evaluating User Simulators and Search Sessions Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Simulators serve two objectives, behavioral realism (matching real user behavior) and tester reliability (producing valid system rankings), and these are often conflated despite being distinct and sometimes conflicting. We present SimEval-IR, an open-source toolkit and benchmark suite that makes this distinction measurable. |
Saber Zerhoudi; |
| 569 | Modalities of Expert Search: How Users Transition Between Names and Topics Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this study, we analyze nearly half a million queries submitted to an institutional expert search system, using a lexicon-based approach to classify each query as name, topic, or unclear. |
Marjan Azimi; Alistair Moffat; Justin Zobel; |
| 570 | ZIPBid: Hierarchical Zero-shot Incremental Spend Planning for Auto-bidding Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As for low-level control, we utilize the Model Predictive Controller (MPC) to track allocations via real-time bid adjustments while strictly adhering to advertiser objectives. |
Yunke Bai; Wenzheng Shu; Jinan Pang; Wentao Bai; Yunshan Peng; Ji Wu; Yanxiang Zeng; Kaiyuan Li; Xialong Liu; |
| 571 | From Continuous Pretraining to Domain-Adaptive Reranking Via Task Vector Adaptation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose a fine-grained task vector adaptation method that learns parameter-wise scaling coefficients for the task vector. |
Sanghyun Cho; Myeongjin Lee; Jong-hun Shin; Jeong Heo; Kiyoung Lee; Soojong Lim; |
| 572 | Reformulating Post-Training As Matrix Factorization for Joint Embedding Refinement Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose REFORM (REformulating post-training as matrix FactOR ization Method), a novel framework that reformulates post-training as a matrix factorization, enabling the joint realignment of user and item embeddings. |
YeoJun Choi; Yoon-Sik Cho; |
| 573 | Delay-Aware Sequential Recommendation with Dynamic Graphs and Variate-Temporal Decomposition Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose DVGD, a delay-aware sequential recommendation framework that learns lag-conditioned message passing over censored histories induced by delayed log availability. |
Yoonhyuk Choi; |
| 574 | FAST-MEL: A Fast, Accurate, and Storage Efficient Solution for Multimodal Entity Linking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we highlight that state-of-the-art systems fail to simultaneously satisfy these 3 requirements. To meet this three-fold objective, we propose FAST-MEL, a lightweight encoder-based MEL solution that relies on a novel and compact fixed-size vectorized representation of both the textual and visual information of each entity or mention. |
Thomas Derrien; Laurent Amsaleg; Pascale S\'{e}billot; |
| 575 | RubricRAG: Towards Interpretable and Reliable LLM Evaluation Via Domain Knowledge Retrieval for Rubric Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we study whether LLMs can generate useful instance-specific rubrics that align with human-authored rubrics and improve response evaluation. |
Kaustubh Dhole; Eugene Agichtein; |
| 576 | Reinforcing User Interest Evolution in Multi-Scenario Learning for Recommender Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This inherent heterogeneity poses a substantial challenge to unified modeling, rendering multi-scenario recommendation a non-trivial task. To address these challenges, we propose RUIE, a novel reinforcement learning-enhanced framework that models dynamic user preference evolution as sequential decision-making problems, and leverages cross-scenario behavior sequences to detect interest shifts and dynamically adjust sample utilization for accurate interest capture. |
Zhijian Feng; Wenhao Zheng; Xuanji Xiao; |
| 577 | Query Performance Prediction Under Corpus Growth in Dense Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we extend the QPP paradigm by studying query performance degradation under corpus inflation in dense retrieval systems. |
Kanishka Ghosh Dastidar; Michael Dinzinger; Laura Caspari; Jelena Mitrovi\'{c}; Michael Granitzer; |
| 578 | On The Robustness of LLM Re-Rankings Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Large language models (LLMs) have become an indispensable part of document ranking systems.In this study we employ two datasets and two LLM models to explore the implications of relevant-item density in the input candidate list, and resilience to the ordering cues provided in the prompt. |
Reyhaneh Goli; Alistair Moffat; |
| 579 | Attend to Fragments: How Key Information Affects Large Language Models for Factual Inconsistency Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Using KIFI, we show that LLMs frequently fail to use the appropriate information to make correct decisions. |
Xindi Guo; Zhen Xie; Patrick H. Chen; |
| 580 | Towards Emotional Intelligence in Conversational AI: How Well Can LLMs Recognise Emotion in Conversations? Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present a systematic evaluation of LLMs for conversational emotion recognition, using human annotation as a gold standard to quantify the degree to which reliance on LLMs may introduce error into both fine-tuning and evaluation workflows. |
Islam A. Hassan; Yvette Graham; |
| 581 | A Multi-Granularity Game-Theoretic Approach to Weakly Supervised Temporal Article Grounding Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, these approaches often face two main drawbacks: (1) They mainly focus on global video-level alignment with the article while neglecting the multi-level relationships between video clips and text hierarchy; (2) They overlook intra-modal feature learning, failing to adequately distinguish salient patterns. To address these issues, we propose a novel multi-granularity game-theoretic method, namely Hierarchical Cooperative Network (HCN), which models video clips and article units (i.e., at word, sentence, and paragraph layers) as players in a game. |
Shuyi He; Hao Wen; Pu Zou; Song Zhou; Qingchao Kong; |
| 582 | Named Entity-Driven Graph Smoothing to Enhance Pretrained Document Embeddings in Clustering Tasks Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: By constructing a document graph based on named-entity similarity, we apply spectral filtering to smooth the embedding matrix, yielding entity-aware representations compatible with any clustering algorithm. |
Imed Keraghel; Mohamed Nadif; |
| 583 | REMICA: Reflective Memory and Interventional Context Alignment with Multi-Agent LLMs for Inappropriate Utterance Detection Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we propose the Reflective Memory and Interventional Context Alignment (REMICA) framework. |
Juyoung Kim; Ji-Hong Park; Sang-Min Choi; Gun-Woo Kim; |
| 584 | Revisiting The Role of Learned Attention Weighting in SASRec Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Across fourteen benchmark datasets, this modification often yields performance comparable to the original model, with clear dataset-dependent exceptions. To explain this heterogeneity, we introduce a stage-wise norm-based decomposition that quantifies self-preserving vs. cross-position mixing within attention blocks. |
Keito Kozaki; Keigo Sakurai; Ren Togo; Takahiro Ogawa; Miki Haseyama; |
| 585 | Semantic Recall for Vector Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce Semantic Recall, a novel metric to assess the quality of approximate nearest neighbor search algorithms by considering only semantically relevant objects that are theoretically retrievable via exact nearest neighbor search. |
Leonardo Kuffo; Ioanna Tsakalidou; Roberta De Viti; Albert Angel; Ji\v{r}\'{\i} I\v{s}a; Rastislav Lenhardt; |
| 586 | PEARL: Profile-based Explainable Agent for Edge LLM Recommendation Via Latent Decomposition Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present PEARL (Profile-based Explainable Agent for edge-model Recommendation via Latent Decomposition), which selects the best-fitting 2B-class edge LLM from a structured user profile and generates a natural-language explanation grounded in interpretable latent alignment dimensions. |
Jinkwon Lee; Hayoung Oh; |
| 587 | Revisiting Poison Pills for Neural Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Experiments across multiple datasets and retrieval approaches show that neural feedback methods achieve higher average effectiveness than lexical methods but are more sensitive to the choice of feedback documents. |
Moriya Menachem; Oren Kurland; Fiana Raiber; |
| 588 | Targeted Parameter Selection for Rank-Biased Measurement Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: The rank-biased family of measurements allows ordered rankings to be compared in a top-weighted manner, with the differential emphasis on early items derived from a decaying geometric distribution across positions in the ranking. In this short paper we consider a range of other weighting functions, including both infinite and finite inverse power law distributions. |
Alistair Moffat; |
| 589 | Temporal User-Agnostic Ranking: Detecting Preference Evolution While Preserving Ethical Principles Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we propose T-UARS (Temporal User-Agnostic Ranking System), a methodological framework that integrates temporal awareness into ranking systems without assigning scores to users. |
Guilherme Ramos; Ludovico Boratto; Mirko Marras; |
| 590 | BANANA: Bounded Adaptive Navigation Architecture for Nested Archives Related Papers Related Patents Related Grants Related Venues Related Experts View Save Abstract: Retrieval-augmented question answering (QA) over enterprise PDF archives frequently produces confident near-miss answers because dense embeddings discard entity identity, … |
Anup Roy; Rishabh Gyanendra Upadhyay; Animesh Rameshbhai Panara; Aidan Philip Millar; Larry Murray; Robin Mills; |
| 591 | BeLink: Biomedical Entity Linking Meets Generative Re-Ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we demonstrate that instruction-tuning of open-source generative models can offer an effective solution when applied at the re-ranking stage of the BEL pipeline. |
Darya Shlyk; Stefano Montanelli; Lawrence Hunter; |
| 592 | Numerical Hallucinations in Retrieval-Augmented Generation: Detection and Analysis Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: With the rise in the usage of Retrieval-Augmented Generation (RAG) systems to improve the factual accuracy of the large language models (LLM), there still exists a concern regarding these systems producing hallucinating outputs not grounded in the retrieved documents. |
Sera Singha Roy; |
| 593 | DisCoRec: Disentangled Conformity-aware Recommendation with LLM-Guided Multi-View Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Even in explicitly disentangled models, attention-based explanations can be unfaithful, as attention weights do not reliably indicate importance. To address these issues, we propose DisCoRec, a multi-view framework that integrates LLM semantics into each view and calibrates view weights via LLM-guided gating. |
Minkyung Song; Soyoung Park; Sungsu Lim; |
| 594 | SAGE: Solver-Aligned Guided Exploration Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose SAGE, a modular approach that trains a small searcher model to handle codebase exploration as a tool for a frozen large solver model. |
Nikita Sorokin; Ivan Sedykh; Timur Ionov; Valentin Malykh; |
| 595 | The LLM Effect on IR Benchmarks: A Meta-Analysis of Effectiveness, Baselines, and Contamination Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We analyze 179 publications reporting on the TREC Robust04 collection and the TREC Deep Learning 2020 (DL20) Passage Retrieval benchmark, using ACM Digital Library keyword search supplemented by citation-graph backtracking for Robust04. |
Moritz Staudinger; Wojciech Kusa; Allan Hanbury; |
| 596 | Clustering-Based Methods for Vector-Based Pseudo-Relevance Feedback Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: As a result, many of these advancements have reduced the ability to utilize token-level statistics. In this work, we aim to explore how well this type of approach can adapt to dense retrieval models when it is not feasible to use surface-form information to pick discriminating expansion tokens. |
Xavier Velez; Andrew Yates; Eugene Yang; Trevor Adriaanse; Sanjeev Khudanpur; |
| 597 | PeaCap: Patch-Level Retrieval for Lightweight Retrieval-Augmented Image Captioning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose PeaCap, a patch-based retrieval-augmented captioning framework that explicitly studies how retrieval granularity affects the quality of retrieved object evidence and downstream caption generation. |
Robin Viltoriano; Wei Emma Zhang; Hu Wang; Mong Yuan Sim; Yanjun Shu; |
| 598 | Teaching Small Models When Not to Call Functions: Structured Reasoning for Tool Refusal in Low-Resource Languages Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose structured reasoning forms: a training-time approach that teaches models to generate explicit reasoning about tool-invocation decisions through lightweight key-value schemas. |
Dung Pham Tuan Vo; Thai Trung Tran; Tushar Semwal; |
| 599 | Prototype-Calibrated Graph Prompting for Few-Shot Graph Adaptation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address the issue, we propose ProtoCalib, a lightweight local graph prompt for few-shot adaptation of frozen GNNs. |
Yumeng Zhao; Yan Yan; Chuang Zhao; Yingzi Shi; Bei Hua; Shuo Wen; |
| 600 | Agentic Multi-Source Grounding for Enhanced Query Intent Understanding: A DoorDash Case Study Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce an Agentic Multi-Source Grounded system that addresses both failure modes by grounding LLM inference in (i) a staged catalog entity retrieval pipeline and (ii) an agentic web-search tool invoked autonomously for cold-start queries. |
Emmanuel Aboah Boateng; Kyle MacDonald; Akshad Viswanathan; Sudeep Das; |
| 601 | How Trendyol Enables Trustworthy A/B Testing at E-Commerce Scale Through Integrated Statistical Methodologies Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we describe the statistical methodology layer developed by Trendyol’s data science teams to improve both the trustworthiness and iteration speed of online experiments at scale. |
Serhat Mirkan Acar; Damla H\i{}zl\i{} Sert; Kaya Tatar; |
| 602 | Cross-Category Basket Recommendation Via Transactional Debiasing and Graph Propagation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Complement discovery must therefore remove popularity artifacts while remaining robust under sparse transactions. We address this setting using large-scale purchase logs as primary evidence.Our approach learns complementary structure from purchase sequences via a debiased transactional scorer and a graph-based completion mechanism. |
Serhat Mirkan Acar; |
| 603 | Negative Data Mining for Contrastive Learning in Dense Retrieval at IKEA.com Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a systematic approach to improving dense retrieval for IKEA product search through structured negative sampling strategies and scalable LLM-as-a-judge relevance evaluation.Building on IKEA Search Engine’s late-interaction retrieval architectures, we introduce two key contributions: (1) structured negative sampling strategies that leverage product hierarchical taxonomy and product attributes to generate semantically challenging negatives, and (2) a comprehensive LLM-based evaluation methodology for generating training data. |
Eva Agapaki; Amritpal Singh Gill; |
| 604 | Beyond Semantic Similarity: Explicit Intent Modeling for Query–Product Matching Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose an aspect-aware ranking framework that retrieves and resolves aspects in queries and performs fine-grained semantic affinity match against aspects in products to compute an aggregate query-product level aspect affinity score. |
Amanuel Alambo; Sathappan Muthiah; Diego Sierra; Zhenzhong Zhang; Atiq Islam; Alex Cozzi; |
| 605 | LLM Agents Factory: Retrieval of Domain-Specific LLM Agents Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, their practical deployment is often limited by the computational cost and instability associated with the on-the-fly agent design for each user request. To address this, we present LLM Agents Factory, a retrieval-based framework that constructs domain-specific and Wikipedia-grounded agents on demand using a base of over 20K predetermined agent profiles. |
Vitalii Belov; Artyom Sosedka; Andrey Sakhovskiy; Elizaveta Kovtun; Artyom Boyarskikh; Semen Budennyy; |
| 606 | RAG-Enhanced Large Language Models for Dynamic Content Expiration Prediction in Web Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Traditional industrial approaches rely on static time-window filtering, resulting in ”one-size-fits-all” rankings where content may be chronologically recent but semantically expired. To address this limitation, we present a novel Large Language Models (LLMs)-based Query Aware Dynamic Content Expiration Prediction Framework deployed in Baidu search, reformulating timeliness as a dynamic validity inference task. |
Tingyu Chen; Wenkai Zhang; Li Gao; Lixin Su; Ge Chen; Dawei Yin; Daiting Shi; |
| 607 | Scaling Search Relevance: Augmenting App Store Ranking with LLM-Generated Judgments Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To maximize relevance, we leverage two complementary objectives: behavioral relevance (results users tend to click or download) and textual relevance (a result’s semantic fit to the query). |
Evangelia Christakopoulou; Vivekkumar Patel; Hemanth Velaga; Sandip Gaikwad; |
| 608 | Unified Supervision for Walmart’s Sponsored Search Retrieval Via Joint Semantic Relevance and Behavioral Engagement Modeling Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose a bi-encoder training framework for Walmart sponsored search retrieval in e-commerce that uses semantic relevance as the primary supervision signal, with engagement used only as a preference signal among relevant items. |
Shasvat Desai; Md Omar Faruk Rokon; Jhalak Nilesh Acharya; Isha Shah; Hong Yao; Utkarsh Porwal; Kuang-Chih Lee; |
| 609 | Agentic Query Reformulation for Contextualized Hyper-Personalized Product Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Traditional e-commerce search often suffers from high abandonment rates due to an intent gap where users provide broad or vague queries. We propose a novel framework using Agentic AI to bridge this gap through dynamic query reformulation. |
Raghav Gaggar; Sean D Rosario; Daniel Varivoda; Shahriar Golchin; Jayant Sachdev; Ali Lafzi; Siddharth Singh; Jason Cho; Yog Domlur; Swati Kirti; Chittaranjan Tripathy; |
| 610 | All The News That Fits in Bits: Learned Rotation-Aware Binary Projections for Efficient News Retrieval at NDTV Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present a two-stage retrieval pipeline deployed at NDTV, one of India’s largest news organizations, that replaces exhaustive float search with a fast binary shortlist followed by a lightweight float rerank. |
Ritwick Ghosh; |
| 611 | What’s in A Name? Product Normalization for Enterprise RAG at NetApp Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This mismatch degrades retrieval precision and recall, ultimately reducing answer quality. I present DOC, the production conversational assistant deployed on docs.netapp.com, and describe two complementary techniques that leverage curated product-name knowledge artifacts to close this vocabulary gap. |
Grant Glass; |
| 612 | A Cascaded Generative Approach for E-Commerce Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This makes it poorly suited to support dynamic objectives and merchandising requirements over time. To address this, we introduce a cascaded merchandising framework that decomposes storefront construction into two generative tasks: (i) placement-level theme generation and (ii) constrained keyword generation per placement to power product retrieval. |
Moein Hasani; Hamidreza Shahidi; Trace Levinson; Yuan Zhong; Guanghua Shu; Vinesh Gudla; Tejaswi Tenneti; |
| 613 | PCANet: Price Change Aware Framework for Mitigating Inconsistencies in Large-Scale Ranking Systems Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we formally define the price change aware ranking problem and propose PCANet, Price Change Aware Framework for Mitigating Inconsistencies in Large-Scale Ranking Systems. |
Maolei Huang; Shuhan Song; Huawei Cao; |
| 614 | SkyDistill: Navigating Fuzzy Flight Search Ranking Via Precise Intent Distillation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Standard ranking models struggle with the domain discrepancy between precise and fuzzy queries, often resulting in a misalignment between offline metrics and online performance. To address these challenges, we propose the Multi-level Cross-scenario Knowledge Distillation (MCKD) framework. |
Maolei Huang; Shuhan Song; Huawei Cao; |
| 615 | A Cost-Efficient AI Copilot for High-Volume E-commerce Customer Support Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present AgentChat Copilot, a human-in-the-loop LLM system deployed at Flipkart, one of the largest e-commerce platforms in India. |
Aditya Jindal; Midthur Ayeshasiddiqa; Adhish Prasoon; |
| 616 | PersonaPlugin: A Multi-Source Persona Framework for LLM Personalization in Telecommunications Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present PersonaPlugin, a production-ready framework that transforms real-world telecom signals—call summaries, app usage, and location patterns—into structured personas for LLM personalization. |
Jinmo Kang; Minseop Lee; Songha Kim; Junho Shin; Changho Lee; Yeonghwan Jeon; Hyuncheol Jo; |
| 617 | Domain-Specific Reranking: When Is Adaptation Worth The Cost? Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we explore domain adaptation of rerankers using unlabeled and labeled, real and synthetic datasets across legal, biomedical, and auditing domains. |
Tuukka Karvonen; Alessio Staffini; Kenichi Maeda; Menaka Arudchelvan; Tomomi Ozawa; Nami Tokunaga; Noriaki Hirokawa; |
| 618 | Constraint Persistence in Conversational Product Search Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose Constraint-Aware Conversational Search (CACS), a framework with three components powered by large language model reasoning: (1) a Constraint Taxonomy Layer that classifies constraints into identity, contextual, and session types through semantic analysis; (2) a Category Transition Layer that determines constraint relevance during category switches using learned compatibility functions; and (3) a Lightweight Persistence Classifier distilled from LLM reasoning for production deployment. |
Saran Kumar Krishnasamy; Inez Wihardjo; |
| 619 | Improving Dating Outcomes By Predicting for Conversations at Hinge Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: By targeting conversations as a late-stage observable proxy for dating intent, this approach captures a higher-fidelity signal than using likes and matches alone. We present the production deployment of this system at Hinge via a multi-task two-tower model that jointly predicts three objectives with a shared embedding layer. |
Frankie Lu; Dan Turkel; Sihan Chen; Behrooz Afghahi; |
| 620 | Building A Production Shopping Agent at Scale Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We share our year-long journey of building a production shopping agent at scale and show that combining agentic reasoning with production-aware retrieval, tool orchestration, and system optimization enables a single shopping agent to support product discovery, shopping question answering, and agentic actions under real-world traffic. |
Chen Luo; Jason Choi; Ziwei Dong; Rahul Dua; Cong Xu; Xuejing Lei; Yuchen Yan; Xin Zhang; Josef Valvoda; Gaurang Sinkar; Binit Jha; Yi Liu; Monica Cheng; |
| 621 | When Search Is Not Enough: Ranking Content for Query-Less Browse Surfaces Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We developed a machine learned browse ranker based on browse session data. |
Shreya Mahapatra; Diksha Bhardwaj; Anandita Chopra; Tracy Holloway King; |
| 622 | From Queries to Playlists: An LLM-Driven Architecture for Semantic Music Search at Scale Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce a semantic playlist generation service that retrieves and ranks tracks based on meaning rather than keyword overlap, while avoiding hallucination-related failures. |
Rinat Mullakhmetov; Fedor Buzaev; Roman Bogachev; Ilya Sedunov; Oleg Pavlovich; Kamil Mazitov; Vladimir Kravtsov; Elena Tutubalina; Daria Pugacheva; Ivan Sukharev; |
| 623 | Policy-Guided RAG: Enforcing Verbatim and Controlled Synthesis Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose a policy-guided transformation pattern for RAG, where retrieved segments carry sensitivity metadata and a lightweight router selects a transformation mode (Verbatim, Combined, or Synthesis). |
Mina Naghash Asadi; Leila Tavakoli; Mustafa Bilgrami; |
| 624 | Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments Via Post-Stratification Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Online evaluation of ranking and retrieval systems often relies on downstream monetization metrics such as app revenue or creator earnings. These metrics are typically heavy-tailed, with a small fraction of users dominating both mean and variance, leading to low statistical power and unreliable conclusions in A/B experiments—especially under limited traffic.We present a practical framework for variance reduction in online experiments by combining post-stratification with CUPED. |
Neeti Pokharna; Olivier Jeunen; Yatharth Saraf; Aleksei Ustimenko; |
| 625 | Pin-SCALE: Semantic Cascading and Alignment Learning for Engagement-Aware IDs in Cold-Start Recommendations Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present Pin-SCALE, a framework for optimized end-to-end integration of SID into dense retrieval in cascading recommender systems. |
Jiaxing Qu; Junpeng Hou; Yijie Ding; Jaewon Yang; Olafur Gudmundsson; Sai Xiao; Huizhong Duan; |
| 626 | Scalable Multi-vector Retrieval for Policy Enforcement on E-commerce Marketplaces Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This paper presents a hybrid enforcement system designed to retrieve relevant content policies for e-commerce product listings. |
Akshit Sarpal; Srinivas Kotamraju; Raviteja Uppalapati; |
| 627 | RAPO: Reason-Aware Preference Optimization for Factual Table Question Answering Over Structured Data Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we introduce Reason-Aware Preference Optimization (RAPO), a novel alignment framework that incorporates reason-augmented preference signals to better guide LLM learning and improve the reliability of structured information retrieval. |
Phaniram Sayapaneni; Musen Wen; Sunil Goda; |
| 628 | Fast and Feasible: Permutation-based Constrained Reranking for Revenue Maximization Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Since solving ILP exactly for every query is slow for deployment to the online service, we propose a lightweight permutation-based reranking approximation algorithm PermR. |
Svetlana Shirokovskikh; Anastasiia Soboleva; Ekaterina Solodneva; Aleksandr Katrutsa; Roman Loginov; Egor Samosvat; |
| 629 | Efficient Multi-Cohort Inference for Long-Term Effects and Lifetime Value in A/B Testing with User Learning Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Consequently, an intervention may appear beneficial in the short term and neutral in the long term while still generating lower total value than the control due to users churn. To address this limitation, we introduce a method that estimates long-term treatment effects (LTE) and residual lifetime value change (Δ ERLV) in short multi-cohort A/B tests under user learning. |
Dario Simionato; Andrea Tonon; Mingxue Wang; Weiguo Wang; Tong Gui; Xiaoyue Li; |
| 630 | Beyond Fluency: Toward Reliable Trajectories in Agentic IR Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: To address compounding error and deceptive fluency, we propose verification gates at each interaction unit and advocate systematic abstention under calibrated uncertainty. |
Anushree Sinha; Srivaths Ranganathan; Debanshu Das; Abhishek Dharmaratnakar; |
| 631 | What Gets Cited: Competitive GEO in AI Answer Engines Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In each trial the two sources differ in exactly one factor; we use brand anonymization and counterbalanced source order to separate content effects from position bias. |
Rahul Vishwakarma; Shushant Kumar; Ratnesh Jamidar; |
| 632 | Efficient LLM Adaptation for Opinion Knowledge Graph Construction: Lessons from The Telecom Industry Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Our goal is to transform noisy, multi-turn forum discussions into Opinion Knowledge Graphs (OKGs) to support downstream retrieval, competitor analysis, and real-time customer service monitoring.We investigate critical engineering trade-offs in the instruction-tuning pipeline, focusing on data efficiency and training strategies. |
Nai-Chi Yang; Yu-Ming Hsieh; Wei-Yun Ma; Kuo-Wei Chang; |
| 633 | LLM-Enhanced Topical Trend Detection at Snapchat Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a large-scale system for identifying emerging topical trends on Snapchat, one of the world’s largest short-video social platforms. |
Hangqi Zhao; Jay Li; Abhiruchi Bhattacharya; Cong Ni; Jason Yeung; Jinchao Ye; Kai Yang; Akshat Malu; Manish Malik; |
| 634 | From Unstructured to Structured: LLM-Guided Attribute Graphs for Entity Search and Ranking Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this paper, we present a two-stage approach combining Large Language Model (LLM)-driven attribute graph construction with graph-aware LLM ranking. |
Yilun Zhu; Nikhita Vedula; Shervin Malmasi; |
| 635 | H-MAPS: Hierarchical Memory-Augmented Proactive Search Assistant for Scientific Literature Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Existing proactive information retrieval systems often suffer from context ambiguity, as they rely solely on on-screen text and ignore the reader’s specific background and intent. In this demonstration, we present H-MAPS (Hierarchical Memory-Augmented Proactive Search Assistant), a proactive literature exploration assistant that resolves this ambiguity by leveraging a three-layered hierarchical memory. |
Koji Nishikawa; Makoto P. Kato; |
| 636 | SLeDoC: System for Legal Document Comparison Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present SLeDoC, a system for pairwise, span-aware semantic document comparison that moves beyond token- or character-level matching to semantic judgments. |
Elisei Rykov; Nikolay Ivanov; Kseniia Petrushina; Maria Bandulevich; Valentin Malykh; Vasily Konovalov; Alexander Panchenko; Ilseyar Alimova; |
| 637 | A Demonstration of WikiRAG: An Evidence-based Link Prediction for Wikidata with Retrieval Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present an interactive demonstration of our WikiRAG (a framework that combines automatic link prediction with retrieval augmented generation) designed for Wikidata link prediction and integrating evidence-based human-in-the-loop validation. |
Rohan Sabu; Ola El Khatib; Djellel Difallah; |
| 638 | ESCOMIC: User Adaptive Explainable Search for Comic Books Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We introduce ESCOMIC, an adaptive and explainable search system for multimodal comic content. |
Suraj Shashidhar; Sayantan Polley; Soumit Roy; Andreas N\{u}rnberger; |
| 639 | Visual RAG at Scale: Tile-Level Spatial Pooling for Efficient Multi-Vector Document Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We present Visual RAG Toolkit, a practical system for scaling visual multi-vector retrieval with training-free, model-aware pooling and multi-stage retrieval. |
Ara Yeroyan; |
| 640 | EasyRAG: A Beginner-Friendly and Interactive Framework for Retrieval-Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: However, for beginners, RAG remains difficult to approach due to its conceptual complexity, computational requirements, and non-trivial system design choices. Motivated by this observation, we propose EasyRAG, a beginner-friendly and interactive framework designed to lower the entry barrier to understanding and experimenting with RAG systems. |
Xuanchen Zhou; Haitao Yu; Kaipeng Li; Yubo Fang; |
| 641 | Towards Building A Standard Benchmark for Low-Resource Nepali Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose the creation of a first standardized Nepali IR benchmark supporting monolingual (Nepali queries rightarrow Nepali documents), cross-lingual (English queries rightarrow Nepali documents), and code-mixed (Nepali-English queries rightarrow Nepali documents) retrieval. |
Praveen Acharya; Bal Krishna Bal; |
| 642 | From Translation to Retrieval: Evaluating LLM-Based Information Retrieval for Hausa and Fongbe Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We propose experiments that compare LLM reranking and query expansion against BM25 and multilingual dense retrieval baselines (mDPR, mContriever) for Hausa and Fongbe using three commercial LLMs. |
Mahounan Pericles Adjovi; Roald Eiselen; Prasenjit Mitra; |
| 643 | Speak Beyond English: Multilingual Prompts Improve Query Classification in Small Language Models Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: With recent advancements, Small language models (SLMs) are increasingly used as preprocessors to handle query classification, routing, and candidate selection in retrieval pipelines, but they are nearly always prompted in English, even when users search in Hindi, Bengali, or code-mixed forms. |
Pratyay Banerjee; Panthadeep Bhattacharjee; Angshuman Jana; |
| 644 | Bridging The Language Gap in Text-to-SQL: Adapting LLMs for Chichewa in A Low-Resource Setting Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we investigate the adaptation of LLMs for Text-to-SQL generation in Chichewa, a low-resource Bantu language spoken by over 12 million people in Malawi and neighboring regions. |
John Emeka Eze; Dunstan Matekenya; Evance Mathewe; |
| 645 | Improving Amharic Information Retrieval with Translative and Multi-Agent Debate Retrieval Augmented Generation Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Retrieval-augmented generation (RAG) has been used to improve the accuracy and transparency of outputs produced by large language models (LLMs) by integrating external knowledge; however, applying RAG to low-resource languages presents unique challenges, including poor embedding representations, low retrieval quality, and semantic gaps caused by the scarcity of digital documents. In this proposal, we address these challenges for a selected low-resource language, Amharic, by using translative and debate-based RAG techniques to improve retrieval and reasoning. |
Abel Jotie; Prasenjit Mitra; |
| 646 | Graph-Enhanced Sentence Retrieval for Multi-Document Summarization in Low-Resource Languages Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: Multi-document summarization for low-resource languages faces a critical trade-off: large language models are computationally prohibitive for most institutions, while smaller models suffer from severe hallucination in abstractive generation. We address this through extractive sentence retrieval, which guarantees faithfulness while operating within constrained computational budgets. |
Xuan-Hung Le; Thi Toan Do; Hoang-Quynh Le; |
| 647 | Detectability and Evaluation Risks of LLM-Generated Answers in Low-Resource Swahili Agricultural Information Retrieval Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This study examines Swahili maize-related question answering as a low-resource agricultural IR task and investigates whether LLM-generated responses are detectably different from human-authored answers. |
Theofrida J. Maginga; Farian S. Ishengoma; |
| 648 | Morphology-Aware Retrieval for Low-Resource Environments: Advancing Information Retrieval for Shona Language Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: In this work, we present a preliminary study of Shona IR using sparse and dense retrieval models, demonstrating significant performance limitations due to morphological complexity and data scarcity. |
Tendai Mukande; Noel O’Connor; Ruvimbo Maud Munetsi; |
| 649 | Offline-First Information Retrieval for Curriculum-Aligned STEM Resources in Low-Resource Namibian Schools Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: We describe the design of an offline-first IR prototype that hosts a curriculum-aligned STEM repository locally and supports keyword search, BM25 ranking, and curriculum-aware facets. |
Josephina Muntuumo; Wilbard Lazarus; Naftali Indongo; |
| 650 | Building Foundations for Information Retrieval in Extremely Low-Resource Ethnic Languages: Evidence from Tanzania Related Papers Related Patents Related Grants Related Venues Related Experts View Save Highlight: This disproportionately affects rural populations (over 65\%), where ECLs are used alongside Kiswahili in daily life. This paper presents the ongoing Ethnic Voices initiative, which addresses this gap by building foundational text and speech resources explicitly relevant to IR research in extremely low-resource multilingual settings. |
Joseph P. Telemala; Onesmo S. Nyinondi; Neema N. Lyimo; Baraka W. Nyamtiga; Farian S. Ishengoma; Wema L. Msigwa; |