# Papers by Affiliation

| Affiliation | Papers |
|---|---|
| **37 Interactive Entertainment** | Tail-Aware Adaptive-k: Query-Adaptive Context Selection for Retrieval-Augmented Generation (TAA-k) |
| **A*STAR** | Think2Go: Generative Next POI Recommendation with LLM Reasoning (Think2Go) |
| **Aampe** | Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments via Post-Stratification (ShareChat/SIGIR 2026) |
| **Aarhus University** | TimeRoute: Time-Aware Modality Routing and Diffusion for Multi-Modal Recommendation (TimeRoute) |
| **ABYAT** | When More Data Is Not Enough: The Context-Sufficiency Frontier in Generative AI Personalization (Context-Sufficiency) — arXiv 2610.00654 |
| **Academia Sinica** | Beyond Uniform Token Training: A Multi-Target Framework for Learning Token-Weighted Objectives in Generative Recommenders (Beyond Uniform Token Training) |
| **ACL 2026** | ReRec, SAGE ‖ Statistically Reliable LLM-Based Ranking Evaluation via Prediction-Powered Inference (PRECISE) ‖ ReasonRec: A Reasoning-Augmented Multimodal Agent for Unified Recommendation (ReasonRec) |
| **Adobe Research** | Multimodal Music Recommendation System using LLMs ‖ Dialogue to Discovery: Attribute-Aware Preference Elicitation for Conversational Product Search Assistants (D2D) |
| **African University of Science and Technology** | Rank-Constrained Deep Matrix Completion for Group Recommendation (Group RC-DMC) |
| **Ahmedabad University** | A Reproducibility Analysis of PO4ISR: Diagnosing and Mitigating Semantic Drift in LLM-Based Session Recommendation |
| **AI VK** | Mitigating Collaborative Semantic ID Staleness in Generative Retrieval ‖ Planning over Matrix-Factorization MDPs for Candidate Generation (MF-MDP Planning) ‖ Long-Term Optimization for Large-Scale Generative Retrieval with Off-Policy REINFORCE ‖ Hierarchical Exponential-Gaussian Mixtures for Watch-Time Distribution Prediction (HEGM) ‖ Scaling Graph Neural Networks for Friend Recommendation: Multi-Hash User Embeddings and Temporal Neighbor Sampling |
| **Aiso** | Purchase Advice and Observable Buyer Responses in Real AI Conversations |
| **Albatross AI** | SwapRec: Warming Up Cold Items Through Training-Time Swaps (SwapRec) |
| **Alibaba** | LoopMemGR: From Behavior Logs to Evolving Memory for Generative Recommendation (LoopMemGR) ‖ SPARC: Sequence-aware Progressive Attribute Routing and Compression Framework for Generative Recommendation (SPARC) ‖ DaV-Gen: End-to-End Generative Retrieval via Draft-and-Verify (DaV-Gen) ‖ From Head to Tail: Asymmetric Knowledge Transfer in Long-tail Recommendation with Generative Semantic IDs ‖ RecGPT-Mobile: On-Device Large Language Models · IntRR: A Framework for Integrating SID Redistribution and Length Reduction · ResRank · Why Users Go There ‖ DeGRe: Dense-supervised Generative Reranking ‖ QGS: Query-Conditioned Generative Search ‖ GPlan: Generative Spatiotemporal Intent Sequence Recommendation (GPlan) ‖ DSIRM: Learning Query-Bridged Discrete Semantic Identifiers for E-commerce Relevance Modeling (DSIRM) ‖ SSRLive: Live Streaming Recommendation with Dynamic Semantic ID (SSRLive) ‖ Differentiable Geometric Indexing for End-to-End Generative Retrieval (DGI) ‖ Macro Graph of Experts for Billion-Scale Multi-Task Recommendation (MGOE) ‖ LoopCTR: Unlocking the Loop Scaling Power for Click-Through Rate Prediction (LoopCTR) ‖ ShopX: A Foundation Model for Intent-to-Item Fulfillment in Agentic Shopping (ShopX) ‖ VENOMREC: Cross-Modal Interactive Poisoning for Targeted Promotion in Multimodal LLM Recommender Systems (VENOMREC) ‖ AgentSelect: Benchmark for Narrative Query-to-Agent Recommendation (AgentSelect) ‖ FORGE: Forming Semantic Identifiers for Generative Retrieval in Industrial Datasets (FORGE) ‖ Prompt Generation Technical Report (PG) ‖ Beyond Semantic IDs: Encoding Business-Value Ranking into Document Identifiers for Generative Retrieval (CRID) ‖ TMallGS: Scaling Unified Feature and Sequence Modeling for Generative E-commerce Search (TMallGS) ‖ RecGPT-V3 Technical Report (RecGPT-V3) ‖ Learning to Forget: Satiation-Aware Long-Sequence Transducers for Mitigating Post-Purchase Redundancy (SAM) ‖ Cheaper is Better: A Discount-Aware Network for Conversion Rate Prediction in E-commerce Recommendation System (DANet) ‖ TSGR: Taobao Search Generative Retrieval (TSGR) ‖ Can Generative Recommendation Reach Cold Items? A Temporal Perspective on Semantic-ID Generation (ColdSID) ‖ TopoGR: Revealing and Preserving Latent Structure of Semantic ID in Generative Recommendation (TopoGR)‖ Learning from the Future: Privileged Self-Distillation for Sequential Recommendation (PSD) ‖ Gwhere: Guess Where You Go — Generative Next Point-of-Interest Recommendation in Amap (Gwhere) ‖ SSR-GRPO: Integrating Supervision and Semantic IDs into Reinforcement Learning for Dense Retrieval in E-commerce (SSR-GRPO) ‖ Cascading Relevance-driven Recommendation Network for CTR Prediction in Trigger-Introduced Recommendation (CRRN) ‖ RecGPT-Mobile-V2 Technical Report ‖ Native Multimodal Representation Learning for Click-Through Rate Prediction in E-Commerce Scenarios ‖ TransRetrieval: Scaling Up Transformer-Based Retrieval for Industrial Recommendation (TransRetrieval) ‖ DCEO: Direct Causal Effect Optimization for Long-Term User Value Modeling in E-commerce Search (DCEO) ‖ PailitaoGR: Latent Think-with-Images for Generative Image Retrieval (PailitaoGR) ‖ Astar: Learning to Propose Evolution Directions for Self-Evolving Industrial AI Systems (Astar) ‖ Generative Retrieval for E-commerce: Jointly Learning Embedding and Codebook with Same Product Cluster ‖ SPAR: Enhancing Industrial-Scale Generative POI Recommendation via Real-World Spatial Perception (SPAR) ‖ LLM4AIGQ: LLM-based AI Guidance Query Generation Framework for Multi Interest Mining (LLM4AIGQ) ‖ HF-SID: High-Fidelity Semantic IDs for Generative Retrieval in Location-Based Services (HF-SID) ‖ Inventory-Grounded Policy-Level Optimization for Training-Free AI Search (IGPO) ‖ VARG: Value-Aware and Ranking-Aligned Generative Retrieval for Dynamic E-commerce Search (VARG) — Taobao & Tmall ‖ Reasoning Quality Matters: Combating Reasoning Collapse in LLM-based Embedding Learning (CoFree) — with Wuhan University & Shanghai Jiao Tong University ‖ IntBMoE: Integrating Block-Level Conditioning into Expert Composition for Full-Participation MoE (IntBMoE) ‖ Beyond Sequences: Distilling Structured Decision Memory for LLM Recommendation (MARI) — arXiv 2610.11501 || TIDE: How Can Recommendation Feedback Evolve Agent Memory? (TIDE) - arXiv 2609.37544  ‖ Don't Rewrite the SID Map: Topology Preservation for Generative Retrieval in Dynamic E-Commerce Search (SID-Map) — arXiv 2610.26105|‖ FineSID: Progressive Learning from Content to Relevance to Personalization for CTR Prediction (FineSID) — arXiv 2610.29138  ‖ SLIDE: Search-to-List Item Distillation for Single-Rollout Generative Retrieval (SLIDE) — arXiv 2610.34764 |
| **Alibaba International Digital Commerce Group** | GRC: Learning to Reflect and Correct (GRC) ‖ LWGR: Lagrangian-Constrained Personalized World Knowledge for Generative Recommendation ‖ SORT: A Systematically Optimized Ranking Transformer for Industrial-scale Recommenders (SORT) ‖ MDGR: A Masked Diffusion Generative Recommendation Framework (MDGR) ‖ SIGMA: A Semantic-Grounded Instruction-Driven Generative Multi-Task Recommender at AliExpress (SIGMA) ‖ EvoRec: Self Evolving Agentic Recommender Systems (EvoRec) ‖ SSR: Beyond Dense Connectivity: Explicit Sparsity for Scalable Recommendation (SSR) ‖ UniSGR: Unified Framework for Semantic ID Generation and Ranking (UniSGR) ‖ GateSID: Adaptive Gating for Balancing Semantic and Collaborative Signals in Recommendation (GateSID) ‖ SAM-D2Q: Aligning Multimodal Doc2Query with Search Demand and Conversion for E-commerce (SAM-D2Q) ‖ EAGER: Enrich-and-Align Generative Query Recommendation from Clicked Items in E-commerce Search (EAGER) ‖ MIMA: Multi-Interest Recommendation via Multi-Positive Exclusive Assignment (MIMA) ‖ LazFormer: Scaling Transformers for Industrial Recommendation via Transferable Generative Pre-training (LazFormer) ‖ Cross-Country Code-Mixing for Generative Recommendation (CMRec) ‖ TacGR (UniSGR-V2): Stepwise Target-Side Feature Anticipation and Crossing for Generative Recommendation (TacGR) — arXiv 2610.16132 |
| **Allegro** | Beyond Co-purchase Relation: Evolution of Complementary Recommendations at Allegro (AlleCompanion) |
| **Allen Institute for AI** | Cold-Start Personalization via Training-Free Priors from Structured World Models (PEP) |
| **Amazon** | RecRec: Latent Interests Recursive Reasoning for Sequential Recommendation (RecRec) ‖ Statistically Reliable LLM-Based Ranking Evaluation via Prediction-Powered Inference (PRECISE) ‖ Bi-NAS: Towards Effective and Personalized Explanation for Recommender Systems via Bi-Level Neural Architecture Search (Bi-NAS) ‖ Continual Low-Rank Adapters for LLM-based Generative Recommender Systems (PESO) ‖ Do LLM Recommenders Know When They're Hallucinating? (Hallucination Auditing) ‖ Improving Item Discoverability in e-Commerce Search via Related Intent Generation ‖ Ask to Be Sure: Informative Interactions for Confident Multi-Turn LLM Recommendation (Ask to Be Sure) ‖ Distill Globally, Adapt Locally: Reasoning Distillation and Product-Type Test-Time Training for Scalable Trade-Up Recommendation ‖ How Calibration Content Shapes Attention-Based Reranking (Interpolated Null Calibration) ‖ Attribute-Conditioned Multimodal Slot Factorization for Controllable Fashion Retrieval (MM-slotgate) ‖ Do Multilingual Encoders Produce Language-Consistent Semantic IDs? (MultiLing-SID) — arXiv 2610.01139 ‖ Efficient Clustering with Quality Guardrails for LLM-based Recommender Systems at Industry Scale — arXiv 2607.19704 |
| **Amazon International Machine Learning** | FOSTER: First-order Dataset Distillation for Text-based Sequential Recommendation |
| **American University** | MURAL: Multimodal Uncertainty-aware Recommendation via Adaptive edge Learning (MURAL) |
| **Anhui University** | ProMax ‖ GraphLoRA: Structure-Aware Low-Rank Adaptation for Large Language Model Recommendation (GraphLoRA)  ‖ BGAFlow: Bezier-Guided Adaptive Flow for Cross-domain Sequential Recommendation (BGAFlow) — arXiv 2610.30825|
| **Ant Group** | BITRec (Modeling Behavioral Intensity and Transitions) ‖ Causal-Invariant Cross-Domain Out-of-Distribution Recommendation (CICDOR) ‖ CoFiRec: Coarse-to-Fine Tokenization for Generative Recommendation (CoFiRec)‖ Generative Recommenders Forget the Long Tail (CAPE) — arXiv 2610.29298 |
| **Apple Inc.** | Fortress ‖ It Takes Two to Match: Co-Evolving Generative Retriever with Reinforcement Learning (CoGR) |
| **Arizona State University** | MetaPlate: Counterfactual-Guided RAG-LLM Tool for Personalized Food Recommendation and Hyperglycemia Prevention  ‖ DPE-Rec: Diversity-Preserving and Efficient Diffusion Language Model Recommendation for Bundle Construction (DPE-Rec) — arXiv 2610.34700 |
| **Artefact Research Center** | A Systematic Study of Semantic ID Spaces for Generative Information Retrieval (Semantic-ID Spaces) — arXiv 2610.08732 ‖ Disentangling Paradigm, Identifier, and Decoding in Generative Retrieval (Disentangling Paradigm/Identifier/Decoding) — arXiv 2610.08716 |
| **Avito** | Fast and Feasible: Permutation-based Constrained Reranking for Revenue Maximization (PermR) |
| **Babes-Bolyai University** | Rank-Constrained Deep Matrix Completion for Group Recommendation (Group RC-DMC) |
| **Baidu Inc.** | Learning User-Aware Recall: Personalized Retrieval in Long-Term Conversational Memory (PPRO) ‖ Cold-Starts in Generative Recommendation: A Reproducibility Study (ColdGenRec) ‖ ICEGR: An Intent-Coherent End-to-End Generative Retrieval Framework for E-commerce Search (ICEGR) ‖ Generate to Explore, Select to Exploit: Aligning LLM-based Headline Generation with Personalized Recommendation (GESE) ‖ MuSeR: Scalable Long-sequence Recommendation with Multi-interest Modeling (MuSeR) |
| **Bangsun Technology** | LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation (LBR) |
| **Bay University** | SAGER: Self-Evolving User Policy Skills for Recommendation Agent |
| **Baze University** | Rank-Constrained Deep Matrix Completion for Group Recommendation (Group RC-DMC) |
| **Beihang University** | A2Gen (Action-Aware Generative Sequence Modeling) · LASAR · SynGR: Unleashing the Potential of Cross-Modal Synergy for Generative Recommendation ‖ Beyond Matching: Category-Guided Latent Intent Reasoning for Generative Retrieval in E-Commerce (CaLIR) ‖ Recommendation as Generation: Unifying Personalized Video Generation and Recommendation at Industrial Scale (RaG) ‖ Intuition-Guided Latent Reasoning for LLM-Based Recommendation (IntuRec) ‖ VENOMREC: Cross-Modal Interactive Poisoning for Targeted Promotion in Multimodal LLM Recommender Systems (VENOMREC) ‖ UnpairGR: Unpaired Modality-Agnostic Generative Recommendation (UnpairGR) ‖ Residual Trajectory Distillation for Generative Retrieval (ResTD) ‖ Learning Multiresolution Relevance for Hierarchical Generative Retrieval (RARS) — arXiv 2609.39312 [opensource] ‖ AgentWebRec: Compact Evidence Fusion over the Agent Web for Personalized Recommendation (AgentWebRec) — arXiv 2610.01705 ‖ SemPOI-RL: Aligning LLM Semantic Reasoning for Interpretable Out-of-Town POI Sequential Generation (SemPOI-RL) — Beihang University, arXiv 2608.30399 [opensource] ‖ EASformer: Environment-Action-State Alignment for Industrial Generative Ranking (EASformer) — arXiv 2610.27691 |
| **Beijing Information Science and Technology University** | Beyond Matching: Category-Guided Latent Intent Reasoning for Generative Retrieval in E-Commerce (CaLIR) |
| **Beijing Institute for General Artificial Intelligence** | BAHSD: Bridging the Long-tail Gap via Adaptive Distillation in Black-box Sequential Recommendation (BAHSD) |
| **Beijing Institute of Technology** | Bridging Behavior and Semantics for Time-aware Cross-Domain Sequential Recommendation (BST-CDSR) ‖ Beyond the Beam: Constructive Repair and Candidate Completion for Generative Recommendation (Beyond the Beam)  ‖ Autoregressive Retriever: Improving Query Understanding from Item Feedback for Universal Multimodal Retrieval (ARR) -- arXiv 2610.11666 |
| **Beijing Jiaotong University** | HiLaR: Hierarchical Latent Reasoning for LLM-based Recommendation (HiLaR)  ‖ When Semantics Fall Short: Persona-Guided Cross-Domain Sequential Recommendation with Large Language Models (PerCDR) — arXiv 2610.25221|
| **Beijing Normal University** | HoloRec: Holistic Encoding and Interleaved Reasoning for Generative Recommendation (HoloRec) ‖ CogRec: Structure-Cognitive Fast-and-Slow Reasoning for Generative Recommendation (CogRec) |
| **Beijing University** | From Head to Tail: Asymmetric Knowledge Transfer in Long-tail Recommendation with Generative Semantic IDs |
| **Beijing University of Posts and Telecommunications** | LIFT: Lifecycle-Aware Interaction Factorization for Unified Retrieval and Ranking (LIFT) - arXiv 2610.10556 |
| **BIFOLD & TU Berlin** | Neighborhood-Aware Unlearning for Generative Recommendation (NAU) — arXiv 2610.27232 |
| **Bilibili Inc.** | LeAP: Learnable Adaptive Permutation for Feature Selection in Heterogeneous and Sparse Recommender Systems (LeAP) |
| **Bloomberg** | When and How to Ask: Dynamic Preference Elicitation Strategies for Conversational Recommendation (COPE) |
| **Brno University of Technology** | Do Neural Retrievers Prefer Certain Documents? Evidence of Learned Relevance Priors |
| **ByteDance** | Bridging Passive and Active: Enhancing Conversation Starter Recommendation via Active Expression Modeling (PA-Bridge) ‖ MuChator: Enabling Active Music Discovery via Conversational Music LLMs in Douyin Music ‖ Rec-Distill: An Industrial Distillation Pipeline for Large-Scale Recommendation Models ‖ MERGE: Next-Generation Item Indexing Paradigm for Large-Scale Streaming Recommendation (MERGE) ‖ Farewell to Item IDs: Unlocking the Scaling Potential of Large Ranking Models via Semantic Tokens (TRM) ‖ IAT: Instance-As-Token Compression for Historical User Sequence Modeling (IAT) ‖ OMEGA: Collaborative Memory Augmentation for Generative Recommendation (OMEGA) ‖ STEPS: A Self-Triggered Agentic Push Recommendation System (STEPS) ‖ TM20K: Teacher Retains Full Tokens, Student Merges Efficiently (TM20K) ‖ From Language to Behavior: Scaling Sequence Transformers for Industrial Recommendation Ranking with Rec-Native Designs (ReST) ‖ World Model-Guided Reinforcement Learning via Counterfactual User Engagement Simulation (WMG-RL) ‖ SequenceO1: End-to-End Ultra-Long (100K) Sequence Modeling in Recommendation with Low-Rank Caching (SequenceO1) ‖ OneTrans-V2: Unifying Retrieval, Pre-rank, and Fine-rank with One Transformer in Industrial Recommender (OneTrans-V2) — arXiv 2609.28589 ‖ Generative End-to-end Ad Retrieval at Douyin (GEAR) — arXiv 2609.39327 ‖ Personalization Matters: Long-Horizon Conversation Agent with User-Centric Information in Online Shopping Interactions (Personalization Matters) — arXiv 2610.11375 |
| **Capital Normal University** | Enhancing Group Recommendation with Memory-Augmented Reasoning in LLM Agent (AGR) |
| **Capital One** | From Clicks to Intent: Cross-Platform Session Embeddings with LLM-Distilled Taxonomy for Financial Services Recommendations ‖ REPREC: Representation Driven Parameter-Efficient Recommendation System (REPREC) |
| **Carnegie Mellon University** | RecRec: Latent Interests Recursive Reasoning for Sequential Recommendation (RecRec) ‖ Effective Reinforcement Learning for Agentic Search by Recycling Zero-Variance Queries During Training (with INESC-ID, NOVA Lisbon) ‖ On the Memorization and Generalization of Generative Recommendation (MemGen-GR) ‖ LLM Agents Enable User-Governed Personalization Beyond Platform Boundaries ‖ Fair on the Surface? Benchmarking Hidden-Output Fairness Gaps in LLM Recommenders (FairGap) |
| **Central China Normal University** | PaletteID: Prototype-Composed Semantic Identifiers for Multimodal CTR Prediction (PaletteID) |
| **Central South University** | Echoes in Filter Bubble: Diagnosing and Curing Popularity Bias in Generative Recommenders (Ghost) ‖ Time-Aware Diffusion based on Preference Disentanglement for Generative Recommendation (TDPM) ‖ SimGR: Escaping the Pitfalls of Generative Decoding in LLM-based Recommendation (SimGR) ‖ ReliGRec: Reliability-Oriented LLM-Based Generative Recommendation via User-Risk-Aware Prompt Routing (ReliGRec) |
| **Central University of Finance and Economics** | An LLM-Powered Semantic Alignment Framework for Journal Recommendation |
| **Chinese Academy of Sciences** | BAHSD: Bridging the Long-tail Gap via Adaptive Distillation in Black-box Sequential Recommendation (BAHSD) |
| **Chinese Academy of Sciences (IIE)** | Beyond Patches: Superpixel Token-based Transformers for Attribute-Specific Fashion Retrieval (SuperFashion) (with UCAS) ‖ HoloRec: Holistic Encoding and Interleaved Reasoning for Generative Recommendation (HoloRec) ‖ Denoising Implicit Feedback for Cold-start Recommendation (DIF) ‖ Can We Steer the Black-Box? Controllability-Centric Evaluation of Recommender Systems (CtrlBench-Rec) ‖ CogRec: Structure-Cognitive Fast-and-Slow Reasoning for Generative Recommendation (CogRec) ‖ RGD: Reward Guided Decoding for Generative Recommendation (RGD) ‖ SA²CRQ: Towards Efficient and Generalizable Retrieval (SA²CRQ) ‖ From Agnostic to Specific: Latent Preference Diffusion for Multi-Behavior Sequential Recommendation (FatsMB) — arXiv 2602.23132|
| **Chinese University of Hong Kong (CUHK)** | One Polluted Page Is Enough: Evaluating Web Content Pollution in Generative Recommenders (FORGE) ‖ Generative Archetype-Grounded Item Representations for Sequential Recommendation (GenAIR) (with McGill, Tongji) ‖ Intuition-Guided Latent Reasoning for LLM-Based Recommendation (IntuRec) ‖ LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation (LBR) ‖ World Model-Guided Reinforcement Learning via Counterfactual User Engagement Simulation (WMG-RL) ‖ Re2A: Situated Conversational Recommendation via Rubric-based Preference Reasoning and Alignment (Re2A) |
| **Chinese University of Hong Kong, Shenzhen** | Matryoshka Hash Representations for Model-Aware Compact Semantic Retrieval (MHR) |
| **Chongqing Normal University** | Iterative Semantic Reasoning from Individual to Group Interests for Generative Recommendation with LLMs (ISRF) |
| **Chongqing University** | IBA: Where Reasoning Matters (IBA) ‖ MLTFR · DC4SR ‖ Self-Distilled Reinforcement Learning for Co-Evolving Agentic Recommender Systems (CoARS) ‖ Personalized Communication Skills for Agentic Recommender Systems (AgentCom) ‖ P3Rec: Distilling Prior-Posterior Preference Reasoning for LLM-based Recommendation (P3Rec)  ‖ Structure-Aligned History Learning with Subspace-Structured Semantic IDs for Generative Recommendation (SAHL) — arXiv 2610.25093|
| **Chongqing University of Technology** | P3Rec: Distilling Prior-Posterior Preference Reasoning for LLM-based Recommendation (P3Rec) ‖ Iterative Semantic Reasoning from Individual to Group Interests for Generative Recommendation with LLMs (ISRF) |
| **Cisco Research** | Multimodal Music Recommendation System using LLMs |
| **CISPA Helmholtz Center for Information Security** | Closing the Indexing-Decoding Gap in Multimodal Generative Retrieval via Prefix Retention Optimization (PRO) ‖ Unifying Search and Recommendation in LLMs via Gradient Multi-Subspace Tuning (GEMS) |
| **City University of Hong Kong** | LLM-as-a-Judge for Reliable and Explainable Offline Evaluation in Top-K Recommendation (LLM Judge) ‖ GloRank (From Local Indices to Global Identifiers) · ComeIR · RPORec: Reinforced Preference Optimization for Reasoning-Augmented Recommendations ‖ UFRec: Uncertainty-Guided Future Learning for Sequential Recommendation ‖ When Text-as-Vision Meets Semantic IDs in Generative Recommendation: An Empirical Study ‖ The Best of Both Worlds: Harmonizing Semantic and Hash IDs for Sequential Recommendation (H²Rec) ‖ Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation (MixRAGRec) ‖ Conditional Memory Enhanced Item Representation for Generative Recommendation (ComeIR) ‖ MIRAGE: Manifold-Informed Flow Matching for Sequential Recommendation (MIRAGE) ‖ HRPO: Hierarchical Residual Policy Optimization for Generative Recommendations (HRPO) ‖ HD-Rec: Hierarchical Quantization with Domain-Adaptive Sparse Routing for Generative Cross-Domain Recommendation (HD-Rec) ‖ Empowering Cross-Domain Sequential Recommendation with Hybrid Tokenization and Serial-Parallel Decoding (GenCDSR) ‖ Recommender System as Slow and Fast Thinkers (DS-Frame) ‖ Beyond Modality Harmony: Orthogonal Purification and Topology-Guided MoE for Conflict-Aware Multimodal Recommendation (OrthoRec) |
| **City University of Macau** | Segmentation-Supervised Complexity-Adaptive Recommendation (S2-CAR) ‖ TRUST: Item-Calibrated Interval Evidence for Temporal Session-Based Recommendation (TRUST) ‖ SimGR: Escaping the Pitfalls of Generative Decoding in LLM-based Recommendation (SimGR) |
| **Columbia University** | TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation (TRACER) |
| **Complexity Science Hub Vienna** | Whose Name Comes Up? III: Persona Prompting Effects in LLM-Based Scholar Recommendation |
| **Coolwei AI Lab** | SafeGEO: Understanding Generative Engine Optimization Risks in Recommendation Agents (SafeGEO) |
| **Cornell University** | Credit-assigned Policy Gradient for Early Stage Retrieval in Two-stage Ranking (CA-PG) ‖ Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems (with UVA, Netflix) ‖ Exploring Bottom-Up Clustering for Creating Semantic IDs |
| **Criteo** | RecoAtlas: From Semantic Plausibility to Set-Level Utility in LLM Recommendation Agents ‖ Off-Policy Evaluation for Semantic ID Recommenders: Does the Model's Own Code Hierarchy Help? |
| **CSIRO's Data61** | Beyond Item Order: Temporal Gap Tokenization for Generative Recommendation with Semantic IDs (ChronoSID) ‖ Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation (MMEACR) |
| **CWI** | Beyond Centralization: User-Controlled Federated Recommendations in Practice |
| **Dalhousie University** | CRAMER: Control via Request-Aware Masking for Editing Recommenders (CRAMER) |
| **Dalian University of Technology** | Think2Go: Generative Next POI Recommendation with LLM Reasoning (Think2Go) ‖ RAGR: Review-Augmented Generative Recommendation |
| **Deezer** | Music Playlist Captioning at Scale with Large Language Models |
| **DePaul University** | When LLM-Based User Profiling Adds Value in Production Streaming Recommendation (WhenLLM-Profiling) ‖ Routing Between Generative and Collaborative User Profiles: A Serving-Time Gate for Controllable Novelty — arXiv 2609.39043 ‖ When LLM-Inferred User Context Adds Value in Production Streaming Recommendation (LLM-Inferred-Context) — arXiv 2609.38999 |
| **Dolby Laboratories** | Multimodal Music Recommendation System using LLMs |
| **DoorDash Inc.** | Joint Optimization of Relevance and Engagement in Multi-Task Ranking for E-Commerce with Efficient LLM Supervision ‖ Mind the Gap: Bridging Behavioral Silos with LLMs in Multi-Vertical Recommendations ‖ One Hierarchy, Two Systems: Semantic Product IDs for Discovery-Surface Ranking and Search-Page Query Reformulation |
| **Duke Kunshan University / Duke University** | COPF: An Online Framework for Deployment-Stable Counterfactual Fairness in Evolving Graphs ‖ From Bootstrapping to Sequence Modeling: A Unified Generative Framework for Personalized Landing-Page Modeling (GLAN) ‖ When History Misleads: Asymmetric Margin Supervision for Instruction-Guided LLM Generative Recommendation (AIMS) — arXiv 2610.02600|
| **East China Normal University** | Learning Variable-Length Tokenization for Generative Recommendation (VarLenRec)  ‖ MuSID-KD: Multigranularity Knowledge Distillation for Generative Recommendation with Semantic IDs (MuSID-KD) — arXiv 2610.35218 |
| **East China University of Science and Technology** | TATK: Triple-Aware Top-K Learning with Knowledge-Grounded Verification for LLM-based Sequential Recommendation (TATK) — EMNLP 2026 |
| **East West University** | JobCCC: A Constraint-Aware Conversational Job Recommendation Benchmark and Method (JobCCC) - arXiv 2610.05787 [opensource] |
| **Eastern Institute of Technology Ningbo** | miniReranker: Efficient Multimodal Reranking through Visual Cache Reuse and Interaction Sparsity (with Netmind.ai, LMU Munich) ‖ CompRank: Efficient LLM Reranking via Token-Level Compression and Decoding-Free Scoring (CompRank) |
| **Glance** | HASTE: Hierarchical and Attribute-aware Supervision for Generative Recommendation (HASTE) — arXiv 2610.17368 |
| **Macau Polytechnic University** | RoleMix: Unifying Sequential and Non-Sequential Features via Semantic Tokenization for Post-Click Conversion Rate Prediction (RoleMix) — KDD Cup 2026 Tencent UniRec Challenge, arXiv 2607.22700 |
| **Nanjing University of Science and Technology** | OT-FlowRec: Optimal Transport-Guided Multi-Bridge Flow for Multimodal Sequential Recommendation (OT-FlowRec) — arXiv 2610.16038 |
| **Octen** | Octen-WebGR: Scaling Generative Retrieval to 200 Million Web Pages (Octen-WebGR) — arXiv 2610.35256 |
| **Optum** | From Scores to Safe Sets: Exact Safety-Constrained Decoding for Medication Recommendation (SafeSets) — arXiv 2610.29228 |

| **Qilu University of Technology** | HEM-Med: A Hierarchical Agentic Memory Framework for Safe Multi-Center Medication Recommendation (HEM-Med) — arXiv 2610.20191  ‖ RES-MR: Risk-Aware Reasoning for Explainable and Safe Medication Recommendation (RES-MR) — SIGIR 2026, arXiv 2610.14474|
| **Sakana AI** | QD-Rec: Open-Ended Evolution of Recommendations via Quality-Diversity (QD-Rec) — Sakana AI — [github.com/SakanaAI/QD-Rec](https://github.com/SakanaAI/QD-Rec) |
| **Santa Clara University** | Evolving Semantic Indexes for Generative Retrieval — arXiv 2610.28568 |
| **Sber AI Lab** | PR-entmax: Mitigating Popularity Bias in Sequential Recommendations with Sparse Prior-Regularized Objective (PR-entmax) — arXiv 2610.17823 |
| **University of Bari Aldo Moro** | Decision4Rec: A Jev-Inspired Framework of Decision Models on Knowledge and Active Graphs for Multi-Agent Recommender Systems (Decision4Rec) — arXiv 2610.26035 [opensource] |
| **University of Liverpool** | RoleMix: Unifying Sequential and Non-Sequential Features via Semantic Tokenization for Post-Click Conversion Rate Prediction (RoleMix) — KDD Cup 2026 Tencent UniRec Challenge, arXiv 2607.22700‖ ReGen: Controllable Generative Recommendation via Guided Token Refinement (ReGen) — arXiv 2610.28642 [opensource] |
| **UX Factory** | ARMED: Adaptive Retrieval-Augmented Medication Recommendation (ARMED) — ICTAI 2026, arXiv 2610.16440 |
| **Xi'an Jiaotong-Liverpool University** | PsyBer-Agent: A Psychology-Driven User Agent for Recommender Systems (PsyBer-Agent) — arXiv 2610.13989‖ ReGen: Controllable Generative Recommendation via Guided Token Refinement (ReGen) — arXiv 2610.28642 [opensource] |
| **eBay** | Representation Curriculum: Stagewise Training for Robust Ranking and Allocation ‖ IntentTune: Using User Demand and Personalization to Resolve "Unknown" Query Intents for E-commerce Search (IntentTune) ‖ Stageboost: Recommending Signals Based on Counterfactual Estimation |
| **Education University of Hong Kong** | SAERec: Constructing Fine-grained Interpretable Intents Priors via Sparse Autoencoders for Recommendation (SAERec) ‖ Segmentation-Supervised Complexity-Adaptive Recommendation (S2-CAR) ‖ TRUST: Item-Calibrated Interval Evidence for Temporal Session-Based Recommendation (TRUST) |
| **Emory University** | Two-Sided State-Space Models for Sequential Recommendation with Non-Random Multimodal Review Feedback (TS-SSM) ‖ From Interests to Semantic IDs: Retrieval-Grounded Credit Assignment for Generative Recommendation (Retrieval-Grounded Credit) ‖ Learning Better Reasoning for Generative Recommendation with Semantic IDs (Evo-Rec) |
| **EPFL** | Contextual Scalarisation Thompson Sampling for Multi-Objective Decisions in Public Media (CSTS) ‖ Efficient and Robust Online Learning to Rank in Decentralized Systems (RankGuard) |
| **Estonian Entrepreneurship University of Applied Sciences** | Who Owns the AI Recommendation? A Multi-Industry Empirical Map of Brand Category Ownership Across Large Language Models (LLM Brand) ‖ Repeated Queries Exhaust an LLM's Brand Recommendations but Not Its Sources ‖ System Attribution in LLM Brand Recommendations (Dmitrij Żatuchin) |
| **Free University of Bozen-Bolzano** | Behavior-Mining, Generative Conversations, and Collaborative Advisory: the Future of Travel and Tourism Recommender Systems (TTRS) — arXiv 2610.08232 |
| **Fudan University** | MARS: Multi-rate Aggregation of Recency Signals for Sequential Recommendation across Sparse and Dense Regimes (MARS) ‖ Teach Multimodal Recommendation Model to See via Personalized Visual Extraction and Adaptive Learning (REVEAL) ‖ From Hidden Profiles to Governable Personalization: Recommender Systems in the Age of LLM Agents ‖ ProRL: Effective Reinforcement Learning for Proactive Recommendation (ProRL) ‖ GateDiffInt: Gate-Mediated Controllable Diffusion and Multi-Intent LLM Distillation for User Behavior Modeling (GateDiffInt) ‖ Adaptive Item-based Collaborative Structures via Noise Rescheduling in Diffusion for Generative Recommendation (ANR-DiffRec) ‖ FedHUR: Learning Hierarchical Utility-Guided Client Relations for Personalized Federated Recommendation (FedHUR) ‖ Drift-Aware Continual Tokenization for Generative Recommendation (DACT) |
| **Georgia Institute of Technology** | Agentic Share-of-Search: A Multi-Agent AI System for Competitive Decision-Making in LLM-Mediated E-Commerce (ASoS) |
| **Glasp Inc.** | Selection, Not Salience: The Shape and Limits of Personalization in Social Highlighting |
| **GNucleus AI** | LLM-Based User Personas for Recommendations at Scale |
| **Google** | Beyond Item IDs: Scaling Short-Form-Video Recommendation via Semantic-Native Long Sequence Modeling ‖ LLM-Based User Personas for Recommendations at Scale ‖ Tokens are All You Need: Dual-purpose Semantic IDs for Achieving LLM-Level I/O Efficiency in Recommendation Systems ‖ TokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems (TokenMinds) ‖ Bi-NAS: Towards Effective and Personalized Explanation for Recommender Systems via Bi-Level Neural Architecture Search (Bi-NAS) ‖ Decomposing Staleness in Recommender Systems: A Dual-Filter Framework for Supersession and Decay (SDF) ‖ RecEvolve: A Knowledge-Driven Autonomous Agent System for Recommender Systems (RecEvolve) ‖ Agentic ML Exploration (A-MLE) for Ads Ranking (A-MLE) ‖ Explainable Recommendations at Scale: LLM Rationales for YouTube Music Artist Discovery (YouTube Music Rationales) || CutBCE: A Fused Binary Cross-Entropy Loss for Large-Vocabulary Sequential Recommendation (CutBCE) - arXiv 2610.05559 [opensource]  ‖ UniLang: A Unified Generative Framework for Integrating Machine-Native Symbols into Pretrained LLMs (UniLang) — arXiv 2608.19529 ‖ ReRankID: Reasoning Reranking with Semantic IDs as the Sole Item Representation (ReRankID) — Google / University of Toronto, arXiv 2610.22797 [opensource] || **Google Cloud AI Research** | CompRank: Efficient LLM Reranking via Token-Level Compression and Decoding-Free Scoring (CompRank)  ‖ DuoLM: Fast and Effective Personalized Recommendation with Language Models (DuoLM) — arXiv 2610.24481|
| **Graz University of Technology** | Whose Name Comes Up? III: Persona Prompting Effects in LLM-Based Scholar Recommendation ‖ Quantifying the Engagement Trap: Impact of Short-form Video Recommender Systems on Users with ADHD |
| **Griffith University** | IBA: Where Reasoning Matters (IBA) ‖ Echoes in Filter Bubble: Diagnosing and Curing Popularity Bias in Generative Recommenders (Ghost) ‖ FOSTER: First-order Dataset Distillation for Text-based Sequential Recommendation ‖ When Text-as-Vision Meets Semantic IDs in Generative Recommendation: An Empirical Study ‖ Guiding Federated Graph Recommendation with LLM-encoded knowledge ‖ Self-Distilled Reinforcement Learning for Co-Evolving Agentic Recommender Systems (CoARS) ‖ SimGR: Escaping the Pitfalls of Generative Decoding in LLM-based Recommendation (SimGR) ‖ VaLiDRec: Variable-Length LLM-Aligned Semantic IDs for Generative Recommendation (VaLiDRec) ‖ EPIC: Explicit Posterior Item Conditioning for Semantic ID Diffusion Recommendation (EPIC) ‖ An Efficient and Effective Agentic Group Shilling Attack on Recommender Systems (AGAS) |
| **Guilin University of Electronic Technology** | Beyond Instance-Level Alignment and Uniformity: Semantic Factor Learning for Collaborative Filtering (SaFeAU) |
| **Hangzhou Dianzi University** | PHKT: Personalized Dynamic Hypergraph-enhanced KAN-Transformer for Multi-behavior Sequential Recommendation (PHKT)  ‖ SharpRec: Sharpness-aware Model Merging with Salience Recovery for LLM-based Cross-Domain Sequential Recommendation (SharpRec) — KDD 2026, arXiv 2607.25366|
| **Harbin Institute of Technology** | TIDE: How Can Recommendation Feedback Evolve Agent Memory? (TIDE) - arXiv 2609.37544 |
| **Harvard University** | One Model, Two Markets: Bid-Aware Generative Recommendation (GEM-Rec) |
| **Hefei University of Technology** | Dynamic Spectral Denoising with Global-Context Attention for Multi-Behavior Recommendation (SpectraMB)  ‖ MLLMRec-R1: Incentivizing Reasoning Capability in LLMs for Multimodal Sequential Recommendation (MLLMRec-R1) — arXiv 2603.06243 [opensource]|
| **Hokkaido University** | Residual Dominance as a Structural Account of Last-Item Reliance in Causal Self-Attention Recommenders (Residual Dominance) |
| **Hong Kong Baptist University** | Denoising Implicit Feedback for Cold-start Recommendation (DIF) ‖ Efficient Swing Computation for Retrieval in Large-Scale Recommender Systems (ASC / K-ASC) |
| **Hong Kong Polytechnic University** | Echoes in Filter Bubble: Diagnosing and Curing Popularity Bias in Generative Recommenders (Ghost) ‖ Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation (MixRAGRec) ‖ Time-Aware Diffusion based on Preference Disentanglement for Generative Recommendation (TDPM) ‖ TAP-PER: Beyond Retrieval: Learning Compact User Representations for Scalable LLM Personalization (TAP-PER) ‖ SAERec: Constructing Fine-grained Interpretable Intents Priors via Sparse Autoencoders for Recommendation (SAERec) ‖ Diffusion Language Model for Recommendation (DLMRec) ‖ SimGR: Escaping the Pitfalls of Generative Decoding in LLM-based Recommendation (SimGR) ‖ Re2A: Situated Conversational Recommendation via Rubric-based Preference Reasoning and Alignment (Re2A) — with CUHK & Sichuan University ‖ PCGNet: Unifying Shared and Specific Information for Fashion Matching Recommendations (PCGNet) || ReMem: Multi-memory GRPO for Long-Context Recommendation Agents (ReMem) - arXiv 2609.37311‖ ReGen: Controllable Generative Recommendation via Guided Token Refinement (ReGen) — arXiv 2610.28642 [opensource] ‖ TGR-EvoDAG: A DAG-based Multi-Agent Code Evolution Framework for Generative Recommendation on Tencent (TGR-EvoDAG) — arXiv 2610.32326 |
| **Hong Kong University of Science and Technology** | Hamiltonian Spectral-Temporal Dissipative Dynamics for Sequential Recommendation (HSR) ‖ Knowledge Distillation across Item and Code Spaces for Parallel Generative Recommendation — arXiv 2610.32943 |
| **Hong Kong University of Science and Technology (Guangzhou)** | Expand More, Shrink Less: Shaping Effective-Rank Dynamics for Dense Scaling in Recommendation (RankElastor) ‖ Feedback-Grounded Policy Discovery for Generative Recommendation (Feedback-Grounded Policy Discovery) ‖ Bridging the Topological Divide: Dual-Branch Graph Distillation with Semantic Priors for Robust Cold-Start Recommendation (DualGCN) — Hong Kong University of Science and Technology (Guangzhou), arXiv 2610.19578 || **Honor Device Co., Ltd** | Think Thrice Before Reranking: Multi-perspective Evidence and Reasoning Integration for Text Reranking (MERIT-Rank) |
| **HSE University** | How Well Does Generative Recommendation Generalize? Variable-Length Semantic IDs for Recommender Systems ‖ Bradley-Terry Rankings for Recommender Systems Across Dataset Taxonomies ‖ Long-Term Optimization for Large-Scale Generative Retrieval with Off-Policy REINFORCE ‖ Separating Thinking from Spelling for Efficient Generative Recommendation (Deco) — arXiv 2610.32980 |
| **Huawei** | Towards Fast Domain Adaptation and Fine-Grained User Simulation for Evaluating Conversational Recommender Systems (AdaptSim) ‖ ReCast  · TurboGR ‖ SIDInspector: A Mapping-First Diagnostic Resource for Semantic-ID Tokenizers ‖ RSIR: Can Recommender Systems Teach Themselves? (RSIR) ‖ SIDScope: A Diagnostic Resource for Semantic-ID Interfaces in Generative Recommendation (SIDScope) ‖ Rethinking Item Tokenization in Generative Recommenders: From Fixed Atoms to Semantic Subwords (SST) ‖ Adaptive Item-based Collaborative Structures via Noise Rescheduling in Diffusion for Generative Recommendation (ANR-DiffRec) ‖ Personalized Task Dependency Graphs for Mitigating Signal Erosion in Multi-Task Recommendation (PTDG) ‖ Enabling High-Bandwidth Flash for Generative Recommendation Serving with Write-Aware KV Cache Policy || EvoSkillRec: Skill-Genome Evolution for Recommender Architecture Discovery (EvoSkillRec) - arXiv 2609.34552 ‖ Graph-Informed Semantic IDs (GrIS): Balancing Semantic and Collaborative Signals via Recursive Graph Partition (GrIS) — arXiv 2610.01533 |
| **Huawei Cloud** | RelayGR |
| **Huazhong University of Science and Technology** | PaletteID: Prototype-Composed Semantic Identifiers for Multimodal CTR Prediction (PaletteID) ‖ UFRec: Uncertainty-Guided Future Learning for Sequential Recommendation ‖ MGDiff: Multi-Interest Sequence Recommendation with Masking GNN-Guided Diffusion (MGDiff) ‖ JBM-Diff: Joint Behavior-guided and Modality-coherence Conditional Graph Diffusion Denoising for Multi-Modal Recommendation (JBM-Diff) — arXiv 2604.03654|
| **Hubei University** | Modality-Aware Identity Construction and Counterfactual Structure Learning for ID-Free Multimodal Recommendation (MAIL) |
| **Hunan University** | LIME-Rec: Auditing Semantic Gains in Sequential Recommendation (LIME-Rec) |
| **ICML 2026** | Credit-assigned Policy Gradient for Early Stage Retrieval in Two-stage Ranking (CA-PG) ‖ COPF: An Online Framework for Deployment-Stable Counterfactual Fairness in Evolving Graphs |
| **IIIT Delhi** | Not All Is Lost: Repairing Lossy User Preference States of Personalization Encoders (REPAIR) — arXiv 2610.01270 |
| **IIT-CNR (Pisa)** | Embedding Surgery: Localized Updates for Adaptive Ranking Correction in Dense Retrieval |
| **Imperial College London** | Teach Multimodal Recommendation Model to See via Personalized Visual Extraction and Adaptive Learning (REVEAL) ‖ SG-UMP: Sequence-Guided Universal Multimodal Prioritization Calculation Framework (SG-UMP) || SpeakGR: Learning Semantic IDs Without Forgetting Language Generation (SpeakGR) - arXiv 2609.35430 |
| **Indeed Inc.** | Single-Token Expected-Value Scoring for Cold-Start Candidate Ranking |
| **Independent Researcher** |   A Rank-One Popularity Component in Dot-Product Recommender Scores: Population Theory and Prior-Separation Evidence ‖ OneFeed: A Unified Generative Framework for Feed Content Enhancement and Query Generation (OneFeed) ‖ The Injection Paradox: Brand-Level Suppression in Safety-Trained LLM Recommendations via RAG Context Injection (Injection Paradox) ‖ The Best of the Two Worlds: Harmonizing Semantic and Hash IDs for Sequential Recommendation (H2Rec) ‖ Auditing Return Conditioning as a Control Knob: An Offline Diagnostic for Decision Transformer Recommendation ‖ The Dice Roll Method: A Standardized Protocol for Repeated-Query Auditing of Large Language Model Brand Recommendations ‖ Should I Be Polite to My LLM Relevance Judge? Tone as a Severity Operating-Point Shift ‖ Following the Preference, Missing the Optimum: Compliance Without Optimization in AI Housing Recommendation ‖ Weather- and Location-Aware Agentic Dining Recommendation (Agentic Dining) ‖ MM-VeriRec: Failure-Guided Fusion for Verifiable Agentic Multimodal Recommendation — arXiv 2609.31718 ‖ Do Evidence-Reading Diagnostics Improve Interface Selection in Small LLM Recommenders? — arXiv 2609.37472 |
| **Indian Institute of Science Education and Research (IISER) Bhopal** | Query-Aware Token Budgeting for Efficient Late-Interaction Visual Document Retrieval |
| **Infobip** | Retrieval, Scoring, and Decoding Shape Performance and Stability in LLM-based Conversational Recommendation (CRS-Performance) |
| **Instacart** | Cascaded Generative Approach |
| **Institute of Computing Technology, Chinese Academy of Sciences** | Can LLM Rerankers Predict Their Own Ranking Performance? ‖ PIT: A Dynamic Personalized Item Tokenizer for End-to-End Generative Recommendation (PIT) |
| **Institute of Information Engineering, Chinese Academy of Sciences** | LWGR: Lagrangian-Constrained Personalized World Knowledge for Generative Recommendation ‖ BAHSD: Bridging the Long-tail Gap via Adaptive Distillation in Black-box Sequential Recommendation (BAHSD) ‖ HoloRec: Holistic Encoding and Interleaved Reasoning for Generative Recommendation (HoloRec) ‖ CogRec: Structure-Cognitive Fast-and-Slow Reasoning for Generative Recommendation (CogRec) ‖ RGD: Reward Guided Decoding for Generative Recommendation (RGD) ‖ UniR²: Unifying Generative Recall and Multi-Objective Ranking in a Single Decoder-Only Sequence (UniR²) ‖ SA²CRQ: Towards Efficient and Generalizable Retrieval (SA²CRQ) |
| **Institute of Science and Technology Austria** | TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation (TRACER) |
| **Institute of Software, Chinese Academy of Sciences** | Generalizing Graph Foundation Models via Hyperbolic Retrieval-Augmented Generation (HyRAG) ‖ D3ER: Supporting Multi-Modal Recommendation via Disentangle and Distillation-based Dynamic Ensemble (D3ER) |
| **International University, Vietnam National University HCMC** | Guiding Federated Graph Recommendation with LLM-encoded knowledge |
| **IRISA / Université de Rennes** | REDSI: Addressing the Reproducibility and Evaluation Consistency of Differentiable Search Indexing for Document Retrieval (REDSI) |
| **IRIT, Université de Toulouse** | PDMR: Passage-Driven Multi-ID Document Retrieval (PDMR) |
| **JD.com** | GenRec: A Preference-Oriented Generative Framework for Large-Scale Recommendation ‖ Graph-GRPO: Dependency-Aware Credit Assignment for Generative E-commerce Search Relevance ‖ Adaptive Loss Balancing for Noise-Robust GRPO in Generative Recommendation (AdaGRPO) ‖ HoloRec: Holistic Encoding and Interleaved Reasoning for Generative Recommendation (HoloRec) ‖ OxygenREC-v2: Internalizing Discrimination into Generative Recommendation (OxygenREC-v2) ‖ CogRec: Structure-Cognitive Fast-and-Slow Reasoning for Generative Recommendation (CogRec) ‖ Restoring Collaborative Signals in Semantic-ID Generative Recommendation via Personalized Natural Language ‖ SA²CRQ: Towards Efficient and Generalizable Retrieval (SA²CRQ) ‖ DEGR: Dual Exploration-Driven Generative Re-Ranking (DEGR) |
| **Jilin University** | SAERec: Constructing Fine-grained Interpretable Intents Priors via Sparse Autoencoders for Recommendation (SAERec) ‖ The Best of Both Worlds: Harmonizing Semantic and Hash IDs for Sequential Recommendation (H²Rec) ‖ RPCBench: A Benchmark for Proactive Premise Critique in LLM-based Recommendation (RPCBench) ‖ WIDE: Wildcard Inference with Dynamic Expansion for Cross-Modal Generative Retrieval (WIDE) ‖ RecToolBench: Benchmarking Recommendation-Specific Tool Orchestration under Fuzzy User Intent (RecToolBench) — arXiv 2609.30717 |
| **Jinan University** | Macro Graph of Experts for Billion-Scale Multi-Task Recommendation (MGOE) |
| **Johannes Kepler University Linz** | A2G-DiffRec ‖ Equal Ranking Quality, Different Decisions: Training Order-Consistent LLM Scorers ‖ SwapRec: Warming Up Cold Items Through Training-Time Swaps (SwapRec) ‖ What Price Fairness? Evaluating Energy - Fairness - Accuracy Trade-off in Recommender Systems ‖ SPRIG: Semantic-ID-enhanced Paths for Knowledge Graph-based Generative Recommendation (SPRIG) |
| **Johns Hopkins University** | ORBIT ‖ Beyond Instance-Level Alignment and Uniformity: Semantic Factor Learning for Collaborative Filtering (SaFeAU) ‖ HyperTrace: Hypothesis-Based Preference Tracing for Online LLM Personalization (HyperTrace) |
| **JPMorganChase** | FARE: Deep Reinforcement Learning for Fair Exposure Constrained Uncertainty-Aware Financial Content Personalization (FARE) — arXiv 2609.31890 |
| **KAIST** | Don't Let Bandit Feedback Pull Continual LLM-Recommender Updates Off Target (ABPO) ‖ One Sequential Recommendation Model Pretrained from Synthetic Priors Predicts Multiple Datasets (SRPFN) ‖ On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategies (IIRG) ‖ ItemRAG: Item-Based Retrieval-Augmented Generation for LLM-Based Recommendation (ItemRAG) ‖ Training-Free LLM-Based Recommendation with Post-LLM Item Refinement Using Collaborative Signals (CoRRe) ‖ A Dual-Expert Strategy Integrating LLMs to Mitigate Negative Transfer in Cross-Domain Sequential Recommendation (DuELRec) ‖ RouteRec: Behavior-Guided Sparse Routing for Sequential Recommendation (RouteRec) — arXiv 2609.39007 [opensource] ‖ A Behavioral Trait Leaks into Preferences: Diagnosing Trait Interference in LLM User Simulators (PQA) — arXiv 2609.25572 [opensource] ‖ Dialogue Generation for Conversational Recommendation Using User Search Interactions and Knowledge Graphs (SearCl) — arXiv 2610.25182 |
| **KDD 2026** | The Pitfall of Scaling Up: Uncovering and Mitigating Popularity Bias Amplification in Scaling Transformer-based Recommenders (SPRINT) ‖ LLM-as-a-Judge for Reliable and Explainable Offline Evaluation in Top-K Recommendation (LLM Judge) |
| **Kent State University** | On the Regularization Landscape for the Linear Recommendation Models |
| **King Abdullah University of Science and Technology (KAUST)** | Generative Late-Interaction Embeddings For Visual Document Retrieval (GLIE) |
| **Know-Center Research** | Conversational Recommendation over Live E-Commerce Catalogues with Self-Refreshing Retrieval |
| **KnowFM @ ACL 2026** | Test-Time Training for Zero-Resource Dense Retrieval Reranking (DART) |
| **Korea University** | Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders (KnowSA_CKP) ‖ Continual Low-Rank Adapters for LLM-based Generative Recommender Systems (PESO) ‖ SCoRD: Semantic-Assisted Continual Retriever-Reranker Distillation for LLM-Based Recommendation (SCoRD) ‖ Profiling What Matters: Context-Aware Item Profiles from Large-Scale Metadata for LLM Recommenders (CAIRO) ‖ Beyond Observed Auxiliary Relations: Environment-Conditioned Modeling for Multi-Behavior Recommendation (BOAR) ‖ SAGE: Semantic Attribute Graphs for Multi-Entity Visual Retrieval (SAGE) |
| **Kuaishou** | DualGR: Generative Retrieval with Long and Short-Term Interests Modeling (DualGR) ‖ Feedback-Grounded Policy Discovery for Generative Recommendation (Feedback-Grounded Policy Discovery) ‖ RGD: Reward Guided Decoding for Generative Recommendation (RGD) ‖ LGRID: Interpretable Representation via LLM-Driven Generative Disentanglement for Local-Life Service Recommendation (LGRID) ‖ OneRetrieval ‖ PROMISE · OpenOneRec · OneMall · OneRec-Think · OneRec-V2 · A2Gen · GloRank · GR4AD · AdaSID · UxSID · RPORec ‖ Taiji ‖ RGCD-Rep ‖ OneReason Technical Report ‖ DeRes ‖ AIR ‖ PIT ‖ GEMs ‖ OneLive ‖ OneSearch-V2 ‖ OneBar ‖ DREAM: Dynamic Refinement of Early Assignment Mappings (DREAM) ‖ Denoising Implicit Feedback for Cold-start Recommendation (DIF) ‖ Recommendation as Generation: Unifying Personalized Video Generation and Recommendation at Industrial Scale (RaG) ‖ UniFormer: Efficient and Unified Model-Centric Scaling for Industrial Recommendation (UniFormer) ‖ AgentX: Towards Agent-Driven Self-Iteration of Industrial Recommender Systems (AgentX) ‖ Advancing Model Research in AgentX: Long-Horizon Autonomy for Industrial Recommender Systems (AgentX-Model) — arXiv 2609.30001 ‖ From Bootstrapping to Sequence Modeling: A Unified Generative Framework for Personalized Landing-Page Modeling (GLAN) ‖ IID-Nav: From Extraction to Navigation: Progressive Retrieval with Indirectly Infinite Depth (IID-Nav) ‖ UniMixer: A Unified Architecture for Scaling Laws in Recommendation Systems (UniMixer) ‖ HGenPush: A Heterogeneous Generative Recommendation Architecture for Industrial Push Notification Systems (HGenPush) ‖ Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems ‖ RECAP: Feedback-Driven Streaming Semantic User Profiles for Short-Video Recommendation (RECAP) ‖ UAME: Uncertainty as Remedy: Mitigating Satisfaction Label Bias in Short Video Multi-Objective Ensemble Ranking (UAME) ‖ Grevo: A Unified Generative Recommendation Framework with Evolutionary Item Indexing (Grevo) ‖ RecoReward: Recommender-Guided Multimodal Description Generation for Recommendation (RecoReward) ‖ Multi-Decoder OneRec: Controllable Generative Retrieval for Multi-Objective Industrial Recommendation (Multi-Decoder OneRec) ‖ WhisperRec: Latent Reasoning for Efficient Foundation Recommendation Models (WhisperRec) ‖ PSG: Pair-Space Generation for Efficient Generative Reranking (PSG) ‖ DIRECTOR: Dynamic Index-based Recommendation with Transport-Optimized Retrieval (DIRECTOR) ‖ HRPO: Hierarchical Residual Policy Optimization for Generative Recommendations (HRPO) ‖ SITA: Semantic Interest Tokens for Target-Aware Compression in Long-Sequence Recommendation (SITA) ‖ UniR²: Unifying Generative Recall and Multi-Objective Ranking in a Single Decoder-Only Sequence (UniR²) ‖ UniGD: A Unified Generative-Discriminative Framework for Industrial Retrieval (UniGD) ‖ HD-Rec: Hierarchical Quantization with Domain-Adaptive Sparse Routing for Generative Cross-Domain Recommendation (HD-Rec) ‖ DrEM: Dual-Side Robust Ensemble Ranking from Noisy User Preference Predictions in Video Recommendation (DrEM) ‖ Once Generated, Ranked: End-to-End Generative Slate Recommendation with Unified Semantic-Collaborative IDs (OGR) ‖ From a Static Multi-Level Small Semantic Codebook to a Dynamic Single-Level Large Semantic Codebook for Generative Recommendation ‖ PRQ-KMeans: Projection Residual Quantization for Semantic ID Tokenization (PRQ-KMeans) ‖ TAGR: Temporally Adaptive Generative Recommendation for Industrial Live-Streaming Advertising (TAGR) ‖ SWIM: Step-Wise Integrated Measure for Session-supervised List Evaluation in Generative Re-ranking (SWIM) ‖ HubMixer: Progressive Latent Hub Mixing for Parameter-Efficient Feature Interaction in Recommendation (HubMixer) ‖ DMRL: Document-Mediated Reinforcement Learning for Skill Optimization in Advertising Recommendation (DMRL) ‖ UniRec: Cross-stage Multi-Task Fusion with Preference Alignment for Cascaded Recommender Systems (UniRec) ‖ Addressing Cross-Stage Decoupling of Semantic and Collaborative Signals in Generative Recommendation (SCRec) ‖ Scaling Articulated Rationales for MLLM-based Recommendation (SARA) ‖ KUAISHOU Explorer LLM-Rec Challenge 2026: Reasoning Generative Recommendation — arXiv 2609.39828 ‖ From Agnostic to Specific: Latent Preference Diffusion for Multi-Behavior Sequential Recommendation (FatsMB) — arXiv 2602.23132 ‖ DiffGRM: Diffusion-based Generative Recommendation Model (DiffGRM) — arXiv 2510.21805 [opensource] ‖ AGR: Agentic Generative Retrieval with Tool-Augmented Search (AGR) — arXiv 2610.26831 ‖ OneLatent: Latent Reasoning for Efficient Foundation Recommendation Models (OneLatent) — arXiv 2610.30675|
| **Kyushu University** | Rethinking Semantic Alignment in LLM-Enhanced Collaborative Filtering: A Spectral Decoupling Approach (UniSpecRec) |
| **Leiden University** | Trustworthy Recommendation in the Era of Large Language Models: Opportunities and Challenges ‖ Closing the Indexing-Decoding Gap in Multimodal Generative Retrieval via Prefix Retention Optimization (PRO) ‖ Unifying Search and Recommendation in LLMs via Gradient Multi-Subspace Tuning (GEMS) ‖ Cold-Starts in Generative Recommendation: A Reproducibility Study (ColdGenRec) ‖ Computed, not Memorized: Bit-Factorized Identifiers for Efficient Generative Retrieval (Bit-Factorized) — arXiv 2610.17467  ‖ UserArena: Benchmarking Interactive User Simulation for Long-Horizon Shopping Agents (UserArena) — arXiv 2610.30932 [opensource]|
| **Leiden University / University of Glasgow** | Differentiable Semantic ID for Generative Recommendation (DIGER) |
| **Li Auto Inc.** | Rethinking Sales Lead Scoring with LLM-based Hierarchical Preference Ranking (HPRO) |
| **LinkedIn** | TransX: Scaling Transformer-based Recommendation via Behavioral and Serving Stream Crossings (TransX) ‖ SIREN-RoPE (Learning to Rotate) ‖ Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation (GTI) ‖ Beyond Interleaving: Causal Attention Reformulations for Generative Recommender Systems ‖ TransX Technical Report: Scaling Sequence Transduction for Recommendation (TransX Technical Report) — arXiv 2610.20913 |
| **Lomonosov Moscow State University** | Planning over Matrix-Factorization MDPs for Candidate Generation (MF-MDP Planning) ‖ Scaling Graph Neural Networks for Friend Recommendation: Multi-Hash User Embeddings and Temporal Neighbor Sampling |
| **LUCID Inc.** | Affective Music Recommendation: A Rollout-Based World Model for Offline Preference Optimization (AMRS) |
| **Maastricht University** | Consensus vs. Dissent: Dynamic LLM Modeling of Subjective Preferences in Group Recommenders |
| **Macquarie University / UNSW** | FLR (Factorized Latent Reasoning) · FLR ‖ Causal-Invariant Cross-Domain Out-of-Distribution Recommendation (CICDOR) ‖ Beyond Item Order: Temporal Gap Tokenization for Generative Recommendation with Semantic IDs (ChronoSID) ‖ Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation (MMEACR) ‖ Audio Embedding Models for Music Recommendation (Audio Embedding Music Rec) |
| **MBZUAI** | ChronoID: Infusing Explicit Temporal Signals into Semantic IDs for Generative Recommendation (ChronoID) |
| **McGill University** | Uncertainty and Fairness Awareness in LLM-Based Recommendation Systems ‖ Generative Archetype-Grounded Item Representations for Sequential Recommendation (GenAIR) (with CUHK, Tongji) ‖ Restoring Collaborative Signals in Semantic-ID Generative Recommendation via Personalized Natural Language ‖ TGR-EvoDAG: A DAG-based Multi-Agent Code Evolution Framework for Generative Recommendation on Tencent (TGR-EvoDAG) — arXiv 2610.32326 |
| **Meta Platforms** | Shape Your Feed: An LLM-based Agentic System for Conversational Recommendation (SYF) — arXiv 2608.06632 |
| **Meituan** | UniPolicy: Unified Objective-Specific Policies for Generative Search Advertising (UniPolicy) |
| **Meituan · MBGR** | SIF: Sample Is Feature — Beyond Item-Level, Toward Sample-Level Tokens for Unified Large Recommender Models (SIF) ‖ NONTP: Not Only NTP (NONTP) — arXiv 2607.12277 ‖ Generative Large-Scale Pre-trained Models for Automated Ad Bidding · FLR · MTServe · DIG · RecRM-Bench  · GenLI ‖ Beyond Matching: Category-Guided Latent Intent Reasoning for Generative Retrieval in E-Commerce (CaLIR) ‖ DOS: Dual-Flow Orthogonal Semantic IDs for Recommendation in Meituan (DOS) ‖ Next-Scale Generative Reranking: A Tree-based Generative Rerank Method at Meituan (NSGR) ‖ UnpairGR: Unpaired Modality-Agnostic Generative Recommendation (UnpairGR) ‖ Decoupled Temporal Encoding for Generative Recommendation (DTE) ‖ Think-to-Personalize: Unifying Reasoning and Retrieval for User-Centric Personalized Dense Retrieval (Think-to-Personalize) ‖ Preference Shapes Relevance: Cross-component Hierarchical Semantic Alignment for Personalized Generative Retrieval (CHAP) ‖ UniCon: A Unified Context-Centric Modeling Paradigm for CTR Prediction (UniCon) |
| **MemTensor (Shanghai) Technology** | MemRetriever: Learning to Search, Reflect, and Retrieve from Long-Term Memory (MemRetriever) |
| **Meta** | Generative Reasoning Re-ranker (GR2) · MARS · LLM Retrieval for Stable and Predictable Ad Recommendations ‖ Credit-assigned Policy Gradient for Early Stage Retrieval in Two-stage Ranking (CA-PG) ‖ Memento: Personalized RAG-Style Long-Retention Data Scaling for META Ads Recommendation ‖ LoopFM: Learning frOm HistOrical RePresentations of Foundation Model for Recommendation ‖ Toward User Preference Alignment in LLM Recommendation via Explicit Context Feedback ‖ Synthetic Data from Cross-Domain Events for Large-Scale Recommendation Systems (SCALR) ‖ ChronoID: Infusing Explicit Temporal Signals into Semantic IDs for Generative Recommendation (ChronoID) ‖ Structuring and Tokenizing Distributed User Interest Context for Generative Recommendation (G2Rec) ‖ Token Factory: Efficiently Integrating Diverse Signals into Large Recommendation Models (Token Factory) ‖ RankGraph-2: Lifecycle Co-Design for Billion-Node Graph Learning in Recommendation (RankGraph-2) ‖ A General Framework for Multimodal LLM-Based Multimedia Understanding in Large-Scale Recommendation Systems (Multimodal LLM Rec) ‖ Intuition-Guided Latent Reasoning for LLM-Based Recommendation (IntuRec) ‖ CMSL: Constructive Multi-Sequence Learning for Recommendation Systems (CMSL) ‖ ReasonRec: A Reasoning-Augmented Multimodal Agent for Unified Recommendation (ReasonRec) ‖ Diffusion-GR2: Diffusion Generative Reasoning Re-ranker (Diffusion-GR2) ‖ Real-Time Hard Negative Sampling via LLM-based Clustering for Large-Scale Two-Tower Retrieval ‖ CORAL: An LLM-Native Harness for Production Recommender Systems (CORAL) ‖ Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation (Scaling Laws) ‖ Cold-Start Personalization via Training-Free Priors (PEP) ‖ SCOReD: Student-Aware CoT Optimization for Recommendation Distillation (SCOReD) ‖ Massive Memorization with Hundreds of Trillions of Parameters (VISTA) ‖ On the Memorization and Generalization of Generative Recommendation (MemGen-GR) ‖ Tokenizing Numerical and Embedding Features for LLM RecSys ‖ WHALE: A Scalable Unified Model for Recommendation with Wukong-HSTU Architecture (WHALE) ‖ Probabilistic Residual Learning for Online Recommendations (PRL) ‖ GRACE: Generative Recommender Acceleration Engine for Real-Time Ads Retrieval (GRACE) ‖ ConnectionMind: Leveraging Social Networks and LLMs for Personalized Recommendation at Meta (ConnectionMind) ‖ UniDot: A Unified Network for Sequence Modeling and Feature Interaction in Large-scale Recommendation (UniDot) ‖ An Event is Worth One Token: Event Tokenization for Industrial-scale LLM Recommendation (AMBER) ‖ Difficulty-Aware Semantic-ID Optimization for Generative Recommendation (DASO) ‖ hLLM: Single Pass Decoding for Generative Reranking (hLLM) ‖ Auto-RecSys: Harnessing Autonomous Research Agents for Industry-Scale Recommender System (Auto-RecSys) ‖ Recommendation Retrievers Need Verifiers: Universal Generative Reranking for Sequential Recommendations (Verifiers) ‖ PCap: Personalized Retrieval-Stage Diversity Capping in Facebook Marketplace (PCap) ‖ LIGE-GR: A Smooth Leap from Ranking to Generative Recommendation in the LLM Era (LIGE-GR) ‖ Guiding the coarse levels of semantic IDs makes the fine levels learnable (Guided SID) ‖ DASO: Difficulty-Aware Semantic-ID Optimization (GRPO rollout-allocation) — [re-hit 2026-09-23] ‖ Exploring Forum Post Retrieval with Generative Modeling — arXiv 2609.38646 ‖ RankEvolve: A Reliable Multi-Agent Auto-Research Harness for Evolving Ranking Models (RankEvolve) — arXiv 2609.39551 ‖ Optimizing Effective Training Time for Large-Scale Recommendation Systems — arXiv 2610.02057 ‖ From Valid to Useful: Post-Verification Acquisition for Recursive Self-Improving Recommendation (DA-RSIR) -- arXiv 2610.04302|
| **Meta AI** | Decision-Oriented Recommendation Reranking: An Empirical Study of Jev (Jev) — with University of Rochester, arXiv 2609.40241  ‖ AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned Prefixes (AlignPoison) — arXiv 2610.14487|
| **Miami University** | Learning Decomposed Contextual Token Representations from Pretrained and Collaborative Signals for Generative Recommendation (DECOR) |
| **Michigan State University** | ReasonRec: A Reasoning-Augmented Multimodal Agent for Unified Recommendation (ReasonRec) ‖ BONSAI: Beyond Fixed Depths and Widths: Optimizing Textual Decoding Tries in LLM-based Generative Recommendation (BONSAI) ‖ ConnectionMind: Leveraging Social Networks and LLMs for Personalized Recommendation at Meta (ConnectionMind) ‖ Fair on the Surface? Benchmarking Hidden-Output Fairness Gaps in LLM Recommenders (FairGap) ‖ The Like Trap: Multi-Stage Poisoning against Agents in Similarity-based Recommendation Systems (Like Trap) — arXiv 2609.27155|
| **Microsoft** | CapsID: Soft-Routed Variable-Length Semantic IDs ‖ TAP-PER: Beyond Retrieval: Learning Compact User Representations for Scalable LLM Personalization (TAP-PER) ‖ The Case Against Generation for Retrieval: Discriminative Language Models as Effective Retrievers (Discriminative Retrieval) ‖ FlashTrie: A GPU-Accelerated Constrained Beam Search for Generative Retrieval (FlashTrie) |
| **Microsoft Research** | Dialogue to Discovery: Attribute-Aware Preference Elicitation for Conversational Product Search Assistants (D2D) |
| **Microsoft Research Asia** | From Hidden Profiles to Governable Personalization: Recommender Systems in the Age of LLM Agents |
| **MILA** | Building a privacy-preserving Federated Recommender system for mobile devices ‖ Affective Music Recommendation: A Rollout-Based World Model for Offline Preference Optimization (AMRS) ‖ STORM: Stepwise Token Optimization with Reward-Guided Beam Search (with Paris-Saclay, CNRS, Sorbonne, Air Liquide, Sapienza) |
| **MLSys 2026** | FreeScale |
| **Momo (Hello Group)** | Task-Blind No MORE: Multi-Task Information Flow in Unified Ranking Backbones (MORE) |
| **Monash University** | Prompt Optimization for User Simulation in Conversational Recommender Systems (UserSimulator) |
| **Moscow State University** | Fast and Feasible: Permutation-based Constrained Reranking for Revenue Maximization (PermR) |
| **Nanjing University** | Denoising Implicit Feedback for Cold-start Recommendation (DIF) ‖ Hyperbolic RQ-VAE enhanced Generative Recommendation with Differential-Length Codebook Strategy (HG-Rec) |
| **Nanjing University of Posts and Telecommunications** | Hyperbolic RQ-VAE enhanced Generative Recommendation with Differential-Length Codebook Strategy (HG-Rec) |
| **Nankai University** | Graph-GRPO: Dependency-Aware Credit Assignment for Generative E-commerce Search Relevance ‖ RecoReward: Recommender-Guided Multimodal Description Generation for Recommendation (RecoReward) ‖ FSGR: Mitigating Token Frequency Bias for Fair SID-Based Generative Recommendation (FSGR) |
| **Nanyang Technological University** | Think When Needed: Model-Aware Reasoning Routing for LLM-based Ranking ‖ GCRS: Generative Conversational Recommender System ‖ VENOMREC: Cross-Modal Interactive Poisoning for Targeted Promotion in Multimodal LLM Recommender Systems (VENOMREC) ‖ Mitigating Matthew Effect: Multi-Hypergraph Boosted Multi-Interest Self-Supervised Learning for Conversational Recommendation (HiCore) ‖ GRACE: Bringing GRACE to Recommendation: Fine-Tuning for Sustainable and Accurate Personalization (GRACE) ‖ Beyond a Single Story: Meta-Reviewing Sparse and Incomplete User-generated Contents for Recommendation (MOSAIC) ‖ ReMaRk: Many Right Answers, One Training Example: Removing Responder Noise from GRPO for LLM Recommenders (ReMaRk) — Nanyang Technological University, arXiv 2610.17771 [opensource] ‖ Tackling the Modality Inhibitance Effect with Wavelet-based Modality Integration in Multi-modal Recommendation (MMWave) — Nanyang Technological University, IEEE TKDE 2026 — [github.com/kennethorq/MMWave](https://github.com/kennethorq/MMWave) || **Nara Institute of Science and Technology (NAIST)** | Rethinking Semantic Alignment in LLM-Enhanced Collaborative Filtering: A Spectral Decoupling Approach (UniSpecRec) |
| **National Institute of Informatics** | AgentRecommender: LLM Agents Enable Customizable Recommender Systems on the User Side (AgentRecommender) |
| **National Taiwan University** | Adapting Generative Recommenders for Multi-Turn Interaction (INTEGER) — with Johns Hopkins University (Eugene Yang), arXiv 2610.08136 |
| **National Technical University of Athens** | Reading the Mood: Emotion-Guided Book-to-Music Recommendation via CGANs and LLMs (SAGA-CDR) |
| **National Super Computing Center Tianjin** | Time-Aware Diffusion based on Preference Disentanglement for Generative Recommendation (TDPM) |
| **National Taiwan University** | Rethinking Fairness in LLM-Based Recommender Systems: A Survey ‖ Structure-Preserving Projection for Mitigating Modality Bias in LLM-Based Sequential Recommendation (Modality Bias) — RecSys 2026 ‖ Are We Really Making Progress in Group Recommendation? Unmasking the Tie-Breaking Illusion (Tie-Breaking) — RecSys 2026 ‖ ICICLE: Expanding Retrieval with In-Context Documents (ICICLE) ‖ Beyond Uniform Token Training: A Multi-Target Framework for Learning Token-Weighted Objectives in Generative Recommenders (Beyond Uniform Token Training) |
| **National University of Defense Technology** | SimGR: Escaping the Pitfalls of Generative Decoding in LLM-based Recommendation (SimGR) ‖ Preference Flow Matching with Spectral Factorization for Micro-video Recommendation (PrismRec) |
| **National University of Singapore** | Learning from the Future: Privileged Self-Distillation for Sequential Recommendation (PSD) ‖ Towards Effective Structured Context Modeling for Conversational Recommender Systems via Dual-node Monte Carlo Tree Search (DREAMS) ‖ Self-Evolving Memory for Generative Recommendation (LION) |
| **National Yang Ming Chiao Tung University** | LSF-SR: Latent Semantic Fusion for Sequential Recommendation via Flow-based Conditional Variational Autoencoders (LSF-SR) — arXiv 2609.29815 |
| **Netease Cloud Music** | L2Rec: Towards Dual-View Understanding of LLMs for Personalized Recommendation |
| **NetEase Games (Fuxi AI Lab)** | PHKT: Personalized Dynamic Hypergraph-enhanced KAN-Transformer for Multi-behavior Sequential Recommendation (PHKT) ‖ AutoLR: Automating the Path from Research to Launch Review in Industrial Recommender Systems (AutoLR) |
| **Netflix** | Towards Generalizable and Efficient Large-Scale Generative Recommenders ‖ Rank-GRPO ‖ Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems (with UVA, Cornell) ‖ GenPage: Towards End-to-End Generative Homepage Construction at Netflix (GenPage) ‖ Exp-RSFT: Exponential Reward Weighting for Fine-Tuning Generative Recommenders under Sparse and Noisy Feedback (Exp-RSFT) ‖ GenRec: An LLM-Backed Recommendation Ranker at Netflix (GenRec) ‖ Recommendation Quality and the Concentration of Consumption: Experimental Evidence from Netflix ‖ Safety as a Constraint: Fine-Tuning a LLM Recommender to Explain Itself ‖ The Lifecycle of LLM-as-a-Judge for Large-Scale Recommendation Explanations ‖ From Prompting to Behavioral Alignment: Personalized LLM Judges for Recommendation Evaluation ‖ Personalized LLM Recommendation with Semantic IDs: Autoregressive Beam Search vs. Multi-Token Prediction (Netflix SID Study) — arXiv 2610.15864 |
| **New York University Abu Dhabi** | Understanding AI Provider Recommendations in Local Service Markets |
| **NJUST / Nanjing** | Harmonizing Generative Retrieval (RecoChain) |
| **Nokia** | SAPO: Step-Aligned Policy Optimization for Reasoning-Based Generative Recommendation |
| **North University of China** | CCLRec: Consensus-driven Contrastive Learning for LLM-enhanced Graph Recommendation (CCLRec) |
| **Northeastern University** | Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation (GTI) ‖ SAILRec: Steering LLM Attention to Dual-Side Semantically Aligned Collaborative Embeddings for Recommendation (SAILRec) ‖ Causal Direct Preference Optimization for Distributionally Robust Generative Recommendation (CausalDPO) ‖ LLM Agents Enable User-Governed Personalization Beyond Platform Boundaries ‖ Adapting Knowledge Graphs for Behavior Denoising in Sequential Recommendation (AdaptedKG) ‖ MOTIF: Motivation-guided Topology Inference for Cold-start Multimodal Recommendation (MOTIF) ‖ Preference-Drift-Aware Subsequence Learning and Hierarchical Context Fusion for Long-Sequence Generative Recommendation (PrefDrift) |
| **Northeastern University, China** | Anatomy of a Decision: Uncertainty-aware Hierarchical Intent Learning via Flow Matching for Multimodal Recommendation (UHIFlow) — arXiv 2609.29609 |
| **Northwestern University** | Evaluating Brand Retrieval and Ranking in Large Language Model Recommendations ‖ LSREP: A Longitudinal State-Replay Protocol for Evaluating Conversational Memory, with ICE v2 as an Audited Local-First Architecture (LSREP) — with Boston University |
| **Novelcore / University of Peloponnese** | Quanta: A Self-Contained Python Library for Hybrid Retrieval over Quantised Embeddings, Lexical Indexes, and Knowledge Graphs (Quanta) |
| **NTT Inc.** | What Limits Does Quantization Place on Dense Top-k Retrieval? A Theoretical Study |
| **NUS** | RecRec: Latent Interests Recursive Reasoning for Sequential Recommendation (RecRec) ‖ CFALR: Collaborative Filtering-Augmented Large Language Model for Personalized Fashion Outfit Recommendation (CFALR) ‖ Reasoning over Semantic IDs · CARE ‖ Mixture-of-Experts Knowledge Graph Retrieval-Augmented Generation for Multi-Agent LLM-based Recommendation (MixRAGRec) ‖ Dynamic Spectral Denoising with Global-Context Attention for Multi-Behavior Recommendation (SpectraMB) ‖ Trustworthy Recommendation in the Era of Large Language Models: Opportunities and Challenges ‖ TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation (TRACER) ‖ MERGE: Next-Generation Item Indexing Paradigm for Large-Scale Streaming Recommendation (MERGE) ‖ Intuition-Guided Latent Reasoning for LLM-Based Recommendation (IntuRec) ‖ Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems ‖ EchoRec: Multi-Item Prediction-Empowered Generative Recommendation via Cycle-Consistent Preference Alignment (EchoRec) |
| **NVIDIA** | MTServe |
| **NYU** | LLM Agents Enable User-Governed Personalization Beyond Platform Boundaries |
| **Ohio State University** | RecRec: Latent Interests Recursive Reasoning for Sequential Recommendation (RecRec) ‖ REPREC: Representation Driven Parameter-Efficient Recommendation System (REPREC) |
| **OPPO** | ToolRec: Calibrated Preference Alignment for Query Recommendation in On-Device Assistants (ToolRec) |
| **PayPal AI** | SAGA: Structure-Attended Generative Action Embedding Model that encodes Multi-Surface User Action Sequences (SAGA) |
| **Peking University** | StageCF · DUET: Joint Exploration of User Item Profiles in Recommendation System · Interests Burn-down Diffusion Process (StageCF) ‖ UFRec: Uncertainty-Guided Future Learning for Sequential Recommendation ‖ Denoising Implicit Feedback for Cold-start Recommendation (DIF) ‖ Generative Retrieval via Diffusion Transformer with Metric-Ordered Sequence Training and Hybrid-Policy Preference Optimization (MO-DiT+HPPO) ‖ An LLM-Powered Semantic Alignment Framework for Journal Recommendation ‖ Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation (MMEACR) ‖ HiLaR: Hierarchical Latent Reasoning for LLM-based Recommendation (HiLaR) ‖ STEPS: A Self-Triggered Agentic Push Recommendation System (STEPS) ‖ SA²CRQ: Towards Efficient and Generalizable Retrieval (SA²CRQ) ‖ STAR: Structured Tokenization and Target-Aware Interest Representation for PCVR Prediction (STAR) ‖ P3Rec: Distilling Prior-Posterior Preference Reasoning for LLM-based Recommendation (P3Rec) |
| **Penn State University** | CCLRec: Consensus-driven Contrastive Learning for LLM-enhanced Graph Recommendation (CCLRec) |
| **Phenikaa University** | Continual Graph Memory for Adaptive Recommendation under Intent Drift (CGM-Rec) |
| **Pinterest** |  Deep-learning Causal Retrieval Optimization for Efficient e-commerce Distribution ‖ Long-term User Engagement Optimization through Model-agnostic Downstream Rewards Learning ‖ MESH: Scaling Up Retrieval with Heterogeneous Content Unification (MESH) ‖ Fine-Tuned LLM as a Complementary Predictor Improving Ads System ‖ UniPinRec: Unifying Generative Retrieval and Ranking at Pinterest Scale ‖ LLMs Need Encoders for Semantic IDs Too (PrefixMem) ‖ PinEqualizer: Full Funnel Content Exploration and Debiasing System at Pinterest (PinEqualizer) ‖ PinRec: Unified Generative Retrieval for Pinterest Recommender Systems (PinRec) ‖ PinDCO: Whole-Page Aware Dynamic Creative Optimization at Scale (PinDCO) ‖ Algorithmic Harms Associated with Generative Model-Augmented Recommendation Systems — arXiv 2609.33073 |
| **Polytechnic University of Bari** | TRACE: A Conversational Framework for Sustainable Tourism Recommendation with Agentic Counterfactual Explanations ‖ Popcorn: A Configurable Benchmark for Visual Evidence in Multimodal Movie Recommendation (Popcorn) ‖ Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems |
| **Polytechnic University of Hong Kong** | CFALR: Collaborative Filtering-Augmented Large Language Model for Personalized Fashion Outfit Recommendation (CFALR) ‖ Atomic Intent Reasoning: Bringing LLM Semantics to Industrial Cross-Domain Recommendations (AIR) (with Kuaishou) |
| **POSTECH** | Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders (KnowSA_CKP) ‖ Federated Variational Preference Alignment with Gumbel-Softmax Prior (FedVPA-GP) ‖ From Overlooked to Explored: Recovering Item Relations via Mixture of Perspectives for Sequential Recommendation (PRISM) ‖ TRACER: Balancing Stability-Plasticity-Cognitivity Trilemma for LLM Enhanced Continual Recommendation (TRACER) ‖ GOD: Enhancing Generalization via Deep Grafting for Sequential Recommendation (GOD) ‖ Personalized and Multi-View Representation for Federated Cold-Start Recommendation (PMFRec) |
| **Princeton University** | tau-Rec: A Verifiable Benchmark for Agentic Recommender Systems (with Independent) |
| **Purdue University** | GraphRAG-IRL ‖ Token-Level Advertising ‖ Epistemic Warrant for LLM Recommendations: Characterizing the Basis for Reliance When Ground Truth Is Unavailable ‖ The Like Trap: Multi-Stage Poisoning against Agents in Similarity-based Recommendation Systems (Like Trap) — arXiv 2609.27155|
| **Qilu University of Technology (Shandong Computer Science Center)** | PaperFlow: Profiling, Recommending, and Adapting Across Daily Paper Streams (PaperFlow) |
| **Qingdao University** | QCMP-CL |
| **Qiyuan Lab** | FedCGR: Federated Cross-Domain Generative Recommendation (FedCGR) — CIKM 2026 |
| **Queen Mary University of London** | Learning Sparse Representations of Multimodal Content for Enhanced Cold Item Recommendation (RecSys 2026) |
| **QuintoAndar Group** | LLM-Based Re-Ranking for Real Estate Search |
| **Radio Television Suisse (RTS)** | Contextual Scalarisation Thompson Sampling for Multi-Objective Decisions in Public Media (CSTS) |
| **Renmin University of China** | Towards Fast Domain Adaptation and Fine-Grained User Simulation for Evaluating Conversational Recommender Systems (AdaptSim) ‖ Divergence Meets Consensus (MDCNS) ‖ TDPM ‖ DS-MLP ‖ CaLIR ‖ Do Generative Recommenders Deepen the Information Cocoon? A Closed-Loop Simulation with LLM-powered User Simulators (RecLoop) ‖ LoopCTR: Unlocking the Loop Scaling Power for Click-Through Rate Prediction (LoopCTR) ‖ Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems ‖ CRAMER: Control via Request-Aware Masking for Editing Recommenders (CRAMER) ‖ GenRecEdit: Adapting Model Editing for Generative Recommendation with Cold-Start Items (GenRecEdit)‖ Learning from the Future: Privileged Self-Distillation for Sequential Recommendation (PSD) ‖ OMEGA: Collaborative Memory Augmentation for Generative Recommendation (OMEGA) ‖ TransRetrieval: Scaling Up Transformer-Based Retrieval for Industrial Recommendation (TransRetrieval)  ‖ Think Harder Where It Matters: Item-Level Adaptive Recursion for Sequential Recommendation (LoopRec) — arXiv 2610.30128 [opensource]|
| **Renmin University of China** | RecSoup: Efficient Model Merging Approach for Controllable Multi-Objective Recommendation (RecSoup) — Renmin University of China, CIKM 2026 — [github.com/35798642/RECSOUP](https://github.com/35798642/RECSOUP) |
| **Renmin University of China / Beijing University of Posts and Telecommunications** | LLaDA-Rec: Discrete Diffusion for Parallel Semantic ID Generation in Generative Recommendation ‖ VirtualMLE: A Virtual ML Engineer that Optimizes Sequential Recommenders (VirtualMLE) ‖ PIT: A Dynamic Personalized Item Tokenizer for End-to-End Generative Recommendation (PIT) |
| **Rensselaer Polytechnic Institute** | From "Strings" to "Things" for Personal Knowledge Graphs (PKG-Rec) |
| **RMIT University** | One Pass, Any Order: Position-Invariant Listwise Reranking for LLM-Based Recommendation (InvariRank) ‖ FunnelAudit: Responsibility Auditing in Multi-Route Recommender Systems (FunnelAudit) |
| **Rutgers University** | TCA4Rec: Token-level Collaborative Alignment for LLM-based Generative Recommendation ‖ AgentSelect: Benchmark for Narrative Query-to-Agent Recommendation (AgentSelect) ‖ Probabilistic Residual Learning for Online Recommendations (PRL) ‖ Kernel-Managed Shared Memory for System-Wide Personalization ‖ Do Multilingual Encoders Produce Language-Consistent Semantic IDs? (MultiLing-SID) — with Amazon, arXiv 2610.01139 |
| **Samsung** | Ocean4Rec: Offline LLM-Derived OCEAN Profiles for Request-Time VOD Reranking ‖ CoPersona: Collaborative Persona Graphs for Robust LLM Personalization (CoPersona) ‖ An LLM-powered Agentic Recommendation System for Connected TV Content Discovery |
| **San Jose State University** | X-KGRank: A Knowledge Graph RAG Framework for Explainable Recommendations via Pattern Mining and LLM Re-Ranking (X-KGRank) |
| **SAP** | RecPFN: Prior-Fitted Networks for In-Context-Based Recommendations (RecPFN) |
| **Sapienza University of Rome** | TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation (TRACER) ‖ STORM: Stepwise Token Optimization with Reward-Guided Beam Search (with MILA, Paris-Saclay, CNRS, Sorbonne, Air Liquide) ‖ SURF: Subtractive Updates for Recommender Forgetting (SURF) |
| **Scrunch AI** | From Prompt to Purchase: How AI Brand Recommendations Move Consumers on the Open Web |
| **Seoul National University (SNU)** | ItemRAG: Item-Based Retrieval-Augmented Generation for LLM-Based Recommendation (ItemRAG) ‖ When Does Supervised Fine-Tuning Reduce Instruction Sensitivity? ‖ PRAGMA: Evaluating Personalized Guidance with Memory Alignment in Lifelong Conversations (PRAGMA) ‖ RouteRec: Behavior-Guided Sparse Routing for Sequential Recommendation (RouteRec) — with KAIST, arXiv 2609.39007 [opensource] |
| **Shandong Normal University** | Federated User Behavior Modeling for Privacy-Preserving LLM Recommendation (SF-UBM) |
| **Shandong University** | ThinkGR ‖ SF-UBM ‖ PRO ‖ Do Generative Recommenders Deepen the Information Cocoon? A Closed-Loop Simulation with LLM-powered User Simulators (RecLoop) ‖ Cold-Starts in Generative Recommendation: A Reproducibility Study (ColdGenRec) ‖ GRACE: Bringing GRACE to Recommendation: Fine-Tuning for Sustainable and Accurate Personalization (GRACE) ‖ Token-Level Credit Assignment Optimization for Generative Document Retrieval ‖ Topology-Masked Unified Backbone for Joint Feature Interaction and Multi-Domain Sequence Modeling (MaskRec) ‖ What Makes a Good Semantic ID for Generative Recommendation? A Reproducibility Study (SID-Repro) |
| **Shandong University of Finance and Economics** | Federated User Behavior Modeling for Privacy-Preserving LLM Recommendation (SF-UBM) |
| **Shanghai Artificial Intelligence Laboratory** | PaperFlow: Profiling, Recommending, and Adapting Across Daily Paper Streams (PaperFlow) ‖ SAERec: Constructing Fine-grained Interpretable Intents Priors via Sparse Autoencoders for Recommendation (SAERec) |
| **Shanghai International Studies University** | TAP-PER: Beyond Retrieval: Learning Compact User Representations for Scalable LLM Personalization (TAP-PER) |
| **Shanghai JTU** | MuonRec ‖ DiffCold: A Diffusion-based Generative Model for Cold-Start Item Recommendation (DiffCold) ‖ CompRank: Efficient LLM Reranking via Token-Level Compression and Decoding-Free Scoring (CompRank) ‖ HCGRec: Hint-Conditioned Generative Recommendation with Semantic IDs (HCGRec) ‖ Information-Guided Selective Modality-Interest Alignment for Multimodal Recommendation (AMUR) ‖ DMRL: Document-Mediated Reinforcement Learning for Skill Optimization in Advertising Recommendation (DMRL) ‖ DoPR: Reusable Compressed Document Prefixes for Efficient LLM Reranking (DoPR) ‖ Reasoning Quality Matters: Combating Reasoning Collapse in LLM-based Embedding Learning (CoFree) |
| **Shanghai University of Finance and Economics** | Aligning Performance with Contribution: Towards Contribution-Aware Fair Recommendation (CPFR) — with Singapore University of Technology and Design (Zhu Sun), arXiv 2610.08245 |
| **ShareChat** | Distributional Approximate Nearest Neighbour Search for Uncertainty-Aware Retrieval (DINOSAUR) ‖ Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments via Post-Stratification (ShareChat/SIGIR 2026) |
| **Shenzhen Technology University** | UFRec: Uncertainty-Guided Future Learning for Sequential Recommendation |
| **Shenzhen University** | EvoReason: Self-Evolving Reasoning Primitive-Guided On-Policy Distillation for Latent Reasoning in Generative Recommendation (EvoReason) ‖ HSUGA ‖ DrEM: Dual-Side Robust Ensemble Ranking from Noisy User Preference Predictions in Video Recommendation (DrEM) ‖ GroundedGEO: Auditing the Evidence Gap in Generative Search Rankings (GroundedGEO) — arXiv 2609.25189 ‖ CALMRec: Causally Aligned Language Memory for Long-Horizon Recommendation — arXiv 2607.23647 |
| **Shopee** | UniRec ‖ Decoupled Residual Quantization for Robust Semantic IDs in Recommendation (DRQ) · OnePiece ‖ Let the Agent Steer: Closed-Loop Ranking Optimization via Influence Exchange (Sortify) ‖ ManCAR: Manifold-Constrained Latent Reasoning with Adaptive Test-Time Computation for Sequential Recommendation (ManCAR) |
| **Shopify** | T-RoPE: Time-Aware Rotary Position Embedding for Sequential Recommendation (T-RoPE) ‖ Revisiting Generative vs. Dense Retrieval: A Static Value Prior Reverses the Gap (Static Value Prior) — arXiv 2610.19065 |
| **Sichuan University** | Towards Effective Structured Context Modeling for Conversational Recommender Systems via Dual-node Monte Carlo Tree Search (DREAMS) ‖ Re2A: Situated Conversational Recommendation via Rubric-based Preference Reasoning and Alignment (Re2A) |
| **SIGIR 2026** | MARC, Rethinking Semantic Collaborative Integration, A2Gen, CARE, PAD-Rec, InvariRank, GenRec, L2Rec ‖ ACE: Anisotropy-Controllable Embedding for LLM-enhanced Sequential Recommendation ‖ FAVE: Flow-based Average Velocity Establishment for Sequential Recommendation ‖ Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders (KnowSA_CKP) ‖ Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments via Post-Stratification (ShareChat/SIGIR 2026) ‖ Beyond Item IDs: Scaling Short-Form-Video Recommendation via Semantic-Native Long Sequence Modeling |
| **Simulacra Research** | Variance Reduction for Heavy-Tailed Monetization Metrics in Ranking Experiments via Post-Stratification (ShareChat/SIGIR 2026) |
| **Singapore Management University** | CFALR: Collaborative Filtering-Augmented Large Language Model for Personalized Fashion Outfit Recommendation (CFALR) ‖ Dual-Diffusional Generative Fashion Recommendation (DualFashion) ‖ Dynamic Spectral Denoising with Global-Context Attention for Multi-Behavior Recommendation (SpectraMB) ‖ Towards Effective Structured Context Modeling for Conversational Recommender Systems via Dual-node Monte Carlo Tree Search (DREAMS) |
| **Singapore University of Technology and Design** | Causal-Invariant Cross-Domain Out-of-Distribution Recommendation (CICDOR) ‖ Think When Needed: Model-Aware Reasoning Routing for LLM-based Ranking |
| **SK Telecom** | Don't Let Bandit Feedback Pull Continual LLM-Recommender Updates Off Target (ABPO) |
| **Snap Inc.** | Masked Diffusion for Generative Recommendation (MaskGR) ‖ SnapLGR: LLM-Based Generative Retrieval for Snapchat Content Recommendation (SnapLGR) ‖ Implicit Reasoning for Large Language Model-based Generative Recommendation (PauseRec) ‖ On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategies (IIRG) ‖ BONSAI: Beyond Fixed Depths and Widths: Optimizing Textual Decoding Tries in LLM-based Generative Recommendation (BONSAI) ‖ CAMIE: Co-Engagement-Aware Multimodal Item Embeddings for Snap Dynamic Product Ads Retrieval (CAMIE) || GRP: Snap's Generative Recommendation Paradigm (GRP) - arXiv 2609.36688 ‖ LLM-based User-Product Relevance Evaluation for Dynamic Product Advertising (DPA Relevance) — Snap Inc., CIKM 2026, arXiv 2610.14550 |
| **Sony Group Corporation** | Overview and Analysis of the RecSys Challenge 2026: Conversational Music Recommendation (RecSys Challenge 2026) — arXiv 2609.33045 |
| **Sony Research India** | RecRec: Recursive Refinement for Sequential Recommendation‖ ATLAS: Learning to Recommend Across Unseen Domains (ATLAS) — arXiv 2608.03899 |
| **Soongsil University** | Do All Nodes Benefit Equally from Knowledge Graphs? Adaptive Node-Aware KG Fusion for Recommendation (AdaKG) |
| **Sorbonne** | STORM: Stepwise Token Optimization with Reward-Guided Beam Search (with MILA, Paris-Saclay, Air Liquide, Sapienza) |
| **South China Agricultural University** | Mitigating Matthew Effect: Multi-Hypergraph Boosted Multi-Interest Self-Supervised Learning for Conversational Recommendation (HiCore) |
| **South China University of Technology** | Think2Go: Generative Next POI Recommendation with LLM Reasoning (Think2Go) |
| **Southeast University** | DeepInterestGR: Mining Deep Multi-Interest Using Multi-Modal LLMs · Deep Interest Mining · Give the Long-tail More SPACE: Promoting Provider Fairness in Next POI Recommendation (SPACE) || LIFT: Lifecycle-Aware Interaction Factorization for Unified Retrieval and Ranking (LIFT) - arXiv 2610.10556 |
| **Southern University of Science and Technology** | GenCAR: Generative Counterfactual Alignment with Risk-Controlled Selection for Out-of-Distribution Recommendation (GenCAR) |
| **Southwest Jiaotong University** | Self-Distilled Reinforcement Learning for Co-Evolving Agentic Recommender Systems (CoARS) |
| **Spotify** |  Causal Representation Learning for Generalisable Recommendation ‖ GLIDE: Deploying Semantic ID-based Generative Retrieval for Large-Scale Podcast Discovery at Spotify (GLIDE) ‖ NEO: A Unified Language Model for Large Scale Search, Recommendation, and Reasoning (NEO) ‖ Hypothesis-Driven Shelf Generation for Personalised Recommendation ‖ Do Sequential Recommendation Benchmarks Really Require Higher-Order Sequence Modelling? (SeqBench) ‖ The Disconnect Between Better Descriptive Reasoning Trace Quality and Recommendation Effectiveness ‖ Who Are We Recommending To? Recommender Systems in the Agentic Web (Agentic Web) ‖ Robust Fusion of Semantic and Behavioural Signals for LLM Reranking in Personalised Search (Robust Fusion) — USRW Workshop @ RecSys 2026, arXiv 2609.25825 ‖ Bootstrapping Conversational Recommendation Agents At Spotify: Synthetic Data Generation and Self-Improvement Loops — arXiv 2609.30297 || Textual User Taste: Natural-Language Taste Profiles at Spotify Scale (Textual User Taste) - arXiv 2609.35285 |
| **Stanford University** | Diagnosing and Mitigating Retrieval Bottlenecks in LLM-Based Cold-Start Recommendation (LHF) ‖ Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook (Moltbook) ‖ MACS: A Hybrid Multi-Agent Framework for Reliable Conversational E-Commerce Recommendation (MACS) ‖ Token-Level Advertising ‖ FlashVector: Agent for Hierarchical Model Serving Stack Optimization (FlashVector) |
| **Stony Brook University** | CRAB ‖ TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation (TRACER) |
| **Nanyang Technological University** | VirusCascade: Hijacking Collaborative Reflection in LLM-Powered Recommender Agents — arXiv 2609.38270 |
| **Sun Yat-sen University / Nanyang Technological University** | HyCoRec: Hypergraph-Enhanced Multi-Preference Learning for Alleviating Matthew Effect in Conversational Recommendation (HyCoRec) ‖ Mitigating Matthew Effect: Multi-Hypergraph Boosted Multi-Interest Self-Supervised Learning for Conversational Recommendation (HiCore) ‖ FacetCRS: Multi-Faceted Preference Learning for Pricking Filter Bubbles in Conversational Recommender System (FacetCRS)  ‖ CARD: Counterfactual Attention Regulation Diffusion for Diffusion-based Sequential Recommendation (CARD) — arXiv 2601.15673 [opensource]|
| **Sungkyunkwan University** | ACE: Anisotropy-Controllable Embedding for LLM-enhanced Sequential Recommendation ‖ On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategies (IIRG) |
| **Taiyuan University of Technology** | MOTIF: Motivation-guided Topology Inference for Cold-start Multimodal Recommendation (MOTIF) |
| **Technical University of Munich** | TRACE: A Conversational Framework for Sustainable Tourism Recommendation with Agentic Counterfactual Explanations  ‖ Behavior-Mining, Generative Conversations, and Collaborative Advisory: the Future of Travel and Tourism Recommender Systems (TTRS) — arXiv 2610.08232|
| **Tel Aviv University** | Breaking the Information Silo: Semantic Personas for Cross-Domain Recommendation (SPHERE) ‖ Monosemanticity in Recommender Systems || Seeing the Context: Image-Derived Contextual Signals for Recommendation (Seeing the Context) - arXiv 2610.08407 |
| **Telefónica Scientific Research** | RecRec: Latent Interests Recursive Reasoning for Sequential Recommendation (RecRec) |
| **Temple University** | Breaking the Information Silo: Semantic Personas for Cross-Domain Recommendation (SPHERE) |
| **Tencent** | Expand More, Shrink Less: Shaping Effective-Rank Dynamics for Dense Scaling in Recommendation (RankElastor) ‖ HiGR: Efficient Generative Slate Recommendation via Hierarchical Planning · Tencent Advertising Algorithm Challenge 2025: All-Modality Generative Recommendation · Unified Value Alignment for Generative Recommendation in Industrial Advertising (UniVA) ‖ SIREN: Unified Multi-Granularity Semantic Interaction for Multi-Modal Lifelong User Interest Modeling ‖ Skill Is Not Document: A Query-Conditional Benchmark and Two-Stage Retriever for LLM Agent Skill Routing (R3) ‖ NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems (NOVA) ‖ RankUp: Towards High-rank Representations for Large Scale Advertising Recommender Systems (RankUp) ‖ Asymmetric Generative Recommendation via Multi-Expert Projection and Multi-Faceted Hierarchical Quantization (AsymRec) ‖ BARGE: Bridging the Structural Gap — Adapting Autoregressive Generation for Recommendation (BARGE) ‖ Intent-Driven Semantic ID Generation for Grounded Conversational News Recommendation (Intent-Driven SID) ‖ EchoRec: Multi-Item Prediction-Empowered Generative Recommendation via Cycle-Consistent Preference Alignment (EchoRec) ‖ Tlow: Flow-based Item Tokenizer for Recommendation (Tlow) ‖ ProRetrieval: Learning to Orchestrate Hybrid Search via Executable Program Synthesis (ProRetrieval) ‖ TGR: Advancing Industrial Recommendation from Generative-Paradigm Ranking toward Unified Generation and Reasoning (TGR) ‖ One-Step Retrieval Framework for Real-Time Sponsored Search Ads Using Hierarchical Text Representations (ANGLE) ‖ NarraLite: Efficient Multimodal Generative Recommendation with Latent Narrative Reasoning (NarraLite) ‖ KuaFu: Compressing Long User Behavior into Understanding at Billion Scale (KuaFu) ‖ DiffuReason: Bridging Latent Reasoning and Generative Refinement for Sequential Recommendation (DiffuReason) — arXiv 2602.09744 ‖ WeGR: Unified Generative Retrieval for Industrial Search Systems in WeChat (WeGR) — arXiv 2610.16979‖ RecSys Factory: Bounding LLM Agent Autonomy to Decision Points in the Industrial Recommender Lifecycle (RecSys Factory) — arXiv 2608.11241 ‖ TGR-EvoDAG: A DAG-based Multi-Agent Code Evolution Framework for Generative Recommendation on Tencent (TGR-EvoDAG) — arXiv 2610.32326 ‖ From Item Scores to Semantic Tokens: Test-Time Collaborative Adaptation for Generative Recommendation (TCBSA) — arXiv 2610.32098  ‖ LoopGR: Parameter-Efficient Scaling for Generative Recommendation via Recurrent Depth (LoopGR) — arXiv 2610.34157 |
| **Tencent Map** | Revisiting General Map Search via Generative Point-of-Interest Retrieval (GenPOI) |
| **Texas A&M University** | Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems ‖ Closing the Long-Short View Gap in Sequential Recommendation without Cached History |
| **Thakur College of Engineering and Technology** | LSREP: A Longitudinal State-Replay Protocol for Evaluating Conversational Memory, with ICE v2 as an Audited Local-First Architecture (LSREP) |
| **The University of Queensland** | PVTG: Personalized Video Thumbnail Generation with Preference-aware Highlight Retrieval (PVTG) ‖ ProMax ‖ Federated User Behavior Modeling for Privacy-Preserving LLM Recommendation (SF-UBM) ‖ FOSTER: First-order Dataset Distillation for Text-based Sequential Recommendation ‖ Whole-Pool Setwise Reranking with Long-Context Language Models (WP-Setwise / DualEnd) ‖ When Text-as-Vision Meets Semantic IDs in Generative Recommendation: An Empirical Study ‖ Self-Distilled Reinforcement Learning for Co-Evolving Agentic Recommender Systems (CoARS) ‖ Learning User-Aware Recall: Personalized Retrieval in Long-Term Conversational Memory (PPRO) ‖ SIREN (Luring LLMs onto the Rocks): PAIR-Driven Preference Manipulation in Web-RAG Recommenders (SIREN) ‖ VaLiDRec: Variable-Length LLM-Aligned Semantic IDs for Generative Recommendation (VaLiDRec) ‖ Towards Efficient Reasoning in LLM-Based Recommender Systems via Model Merging (REAM) ‖ Personalized Communication Skills for Agentic Recommender Systems (AgentCom) ‖ Empowering Compact LLMs with Fusion of Layer-wise Exits for Recommendation (FLEXRec) ‖ Enhancing Group Recommendation with Memory-Augmented Reasoning in LLM Agent (AGR) ‖ AdaM-Rec: Adaptive Modality Routing for Multimodal Recommendation (AdaM-Rec) — arXiv 2609.38455 [opensource]|
| **The University of Tokyo** | Generative Marketing Mix Modeling: A Causal Inference Framework Linking GEO and GEM to Business Impact (GMMM) |
| **The Walt Disney Company** | AURA: Agentic Diagnosis and Refinement for Production Recommender Systems at Scale (AURA) |
| **Thomson Reuters Labs** | Equal Ranking Quality, Different Decisions: Training Order-Consistent LLM Scorers |
| **Tianjin University** | GCIB: Graph Contrastive Information Bottleneck for Multi-Behavior Recommendation ‖ Feedback-Grounded Policy Discovery for Generative Recommendation (Feedback-Grounded Policy Discovery) ‖ From Feature Interaction to Feature Transport - A Unified Block for Scalable Recommendation Models (CRAFT) |
| **TMLR 2026** | Reproducing FACTER: Fairness via Conformal Thresholding and Prompt Repair |
| **TikTok** | X-Rec Technical Report (X-Rec) — arXiv 2609.29180 || HELIX: Purified and Unified Large-Scale Recommendation (HELIX) - arXiv 2609.37183 |
| **Tongji University** | CFALR: Collaborative Filtering-Augmented Large Language Model for Personalized Fashion Outfit Recommendation (CFALR) ‖ Generative Archetype-Grounded Item Representations for Sequential Recommendation (GenAIR) (with CUHK, McGill) ‖ The Best of Both Worlds: Harmonizing Semantic and Hash IDs for Sequential Recommendation (H²Rec) ‖ Adaptive Item-based Collaborative Structures via Noise Rescheduling in Diffusion for Generative Recommendation (ANR-DiffRec) ‖ CARA: Cognitive Adaptive Recommendation Agent — The Chinese University of Hong Kong / Tongji University, arXiv 2608.16919 |
| **Trine University** | Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems |
| **Tsinghua University** | PVTG: Personalized Video Thumbnail Generation with Preference-aware Highlight Retrieval (PVTG) ‖ Enhancing Local Life Service Recommendation with Agentic Reasoning in Large Language Model · TwiSTAR · AsymRec ‖ VirtualMLE: A Virtual ML Engineer that Optimizes Sequential Recommenders (VirtualMLE) ‖ Generalizing Graph Foundation Models via Hyperbolic Retrieval-Augmented Generation (HyRAG) ‖ The Best of Both Worlds: Harmonizing Semantic and Hash IDs for Sequential Recommendation (H²Rec) ‖ APAO: Bridging the Training-Inference Gap in Generative Recommendation via Adaptive Prefix-Aware Optimization (APAO) ‖ Asymmetric Generative Recommendation via Multi-Expert Projection and Multi-Faceted Hierarchical Quantization (AsymRec) ‖ Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation (MMEACR) ‖ Conditional Memory Enhanced Item Representation for Generative Recommendation (ComeIR) ‖ STAR: Structured Tokenization and Target-Aware Interest Representation for PCVR Prediction (STAR) ‖ Tlow: Flow-based Item Tokenizer for Recommendation (Tlow) ‖ TAGR: Temporally Adaptive Generative Recommendation for Industrial Live-Streaming Advertising (TAGR) ‖ SelfDR: Self-Distillation from Reasoning for LLM-Based Recommendation (SelfDR) ‖ FineSID: Scalable and Efficient Semantic Identifier Learning for Generative Recommendation — arXiv 2609.36670 ‖ FairDiff: Mitigating the Self-Reinforcing Matthew Effect in Diffusion Recommender Models (FairDiff) |
| **TU Delft** | Efficient and Robust Online Learning to Rank in Decentralized Systems (RankGuard) |
| **TU Wien** | VCG: A Multimodal Retrieval Framework for E-Commerce Video Feeds under Extreme Cold-Start Conditions (VCG) |
| **Tubi** | TubiFM: Unified Item, Carousel, and Search Ranking for Streaming Discovery ‖ Bridging the Semantic-Collaborative Gap: An Asymmetric Graph Architecture for Cold-Start Item Recommendation (Shallow-RHS) |
| **Uber AI** | When the Label Ignores the Request: Auditing Policy-Selected Targets in Synthetic Conversational Music Recommendation (When-the-Label) — arXiv 2609.39696 [opensource] |
| **UC Berkeley** | LLM Agents Enable User-Governed Personalization Beyond Platform Boundaries |
| **UC Davis** | Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation (GTI) |
| **UC Riverside** | SCOReD: Student-Aware CoT Optimization for Recommendation Distillation (SCOReD) |
| **UC San Diego** | Expressiveness Limits of Autoregressive Semantic ID Generation in Generative Recommendation (Latte) · F-GRPO  · SID-MLP · FusID: Modality-Fused Semantic IDs for Generative Music Recommendation (FusID) ‖ SafeGEO: Understanding Generative Engine Optimization Risks in Recommendation Agents (SafeGEO) ‖ On the Memorization and Generalization of Generative Recommendation (MemGen-GR) ‖ How Calibration Content Shapes Attention-Based Reranking (Interpolated Null Calibration) ‖ Personal-Agent Mediated Recommendation with Cross-Platform User History (MediateRec) — arXiv 2610.07588 || OpticalRec: Treating Text as Visual Glyphs for Multimodal Recommendation (OpticalRec) - arXiv 2610.05432 || JobCCC: A Constraint-Aware Conversational Job Recommendation Benchmark and Method (JobCCC) - arXiv 2610.05787 [opensource] |
| **UCLA** | LLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender Systems |
| **UIUC Illinois** | Rec-R1 ‖ LRanker: LLM Ranker for Massive Candidates ‖ Structuring and Tokenizing Distributed User Interest Context for Generative Recommendation (G2Rec) ‖ Continual Low-Rank Adapters for LLM-based Generative Recommender Systems (PESO) ‖ Learning Decomposed Contextual Token Representations from Pretrained and Collaborative Signals for Generative Recommendation (DECOR) ‖ LLM Agents Enable User-Governed Personalization Beyond Platform Boundaries ‖ Topology-Aware Tokenization for Generative Recommendation (TopoTok) ‖ Understanding Semantic IDs: From Item Representation to Item Selection in Generative Recommendation (SID Understanding) ‖ Overview of the TREC 2025 Product Search and Recommendation Track ‖ CoFiRec: Coarse-to-Fine Tokenization for Generative Recommendation (CoFiRec)  ‖ RAR: Retrieval Augmented Conversational Recommendation with Reinforcement Learning (RAR) — arXiv 2604.04457 [opensource]|
| **Unbox AI** | Scaling Laws for Behavioral Foundation Models over User Event Sequences |
| **UNIST (Ulsan National Institute of Science and Technology)** | ExPerT: Personalizing LLM Responses to Users' Domain Expertise via Query-Wise Semantic and Keystroke Behavioral Cues (ExPerT) |
| **Universidad Autónoma de Madrid** | Beyond Centralization: User-Controlled Federated Recommendations in Practice  ‖ Behavior-Mining, Generative Conversations, and Collaborative Advisory: the Future of Travel and Tourism Recommender Systems (TTRS) — arXiv 2610.08232|
| **Universidad de Buenos Aires / CONICET** | Do Neural Retrievers Prefer Certain Documents? Evidence of Learned Relevance Priors |
| **University College Dublin / Huawei Ireland Research Centre** | From Raw IDs to Semantic Planning: How Recommender Systems Utilize Information at Scale ‖ RAMP: Robust Ad Recommendation Under Limited Personalized-Feature Availability via Masking and Alignment Pathways (RAMP) ‖ PriCoRec: A Privacy-Aware Cloud-Device Collaborative Framework for Ad Recommendation under Feature Constraints (PriCoRec) |
| **University of Amsterdam** | Closing the Indexing-Decoding Gap in Multimodal Generative Retrieval via Prefix Retention Optimization (PRO) ‖ Reproducing FACTER: Fairness via Conformal Thresholding and Prompt Repair ‖ Cold-Starts in Generative Recommendation: A Reproducibility Study (ColdGenRec) ‖ TimeRoute: Time-Aware Modality Routing and Diffusion for Multi-Modal Recommendation (TimeRoute) ‖ LSNR: Learned Sparse News Recommendations with Attentive User Modeling (LSNR) — University of Amsterdam, SIGIR-AP 2026 || **University of Applied Sciences Ravensburg-Weingarten** | POI Recommendation with LLM-Augmented Multi-Graph Learning and Contrastive Alignment (LLM-MGCL) |
| **University of Arizona** | Fairness Attacks on Recommender Systems |
| **University of Bari / University of Glasgow** | URecJPQ: Memory-efficient Multimodal Recommendation Models through RecJPQ in Large-Scale Scenarios (URecJPQ) |
| **University of Birmingham** | RouteRec: Strict Evaluation of Recommender-Agent Selection and Aggregation (RouteRec) |
| **University of British Columbia** | CompRank: Efficient LLM Reranking via Token-Level Compression and Decoding-Free Scoring (CompRank) ‖ Recommendation World Models for Future-State Control (UA-TWM) |
| **University of California, Berkeley** | Trading Engagement for Sustainability: Carbon-Aware Re-ranking for E-commerce Recommendations ‖ Dialogue to Discovery: Attribute-Aware Preference Elicitation for Conversational Product Search Assistants (D2D) |
| **University of Catania / CNR** | RAGEAR: Retrieval-Augmented Graph-Enhanced Academic Recommender |
| **University of Chinese Academy of Sciences** | BRIDGE: Behavior-Guided Candidate Calibration for Multimodal Recommendation ‖ BAHSD: Bridging the Long-tail Gap via Adaptive Distillation in Black-box Sequential Recommendation (BAHSD) ‖ Generalizing Graph Foundation Models via Hyperbolic Retrieval-Augmented Generation (HyRAG) ‖ Can LLM Rerankers Predict Their Own Ranking Performance? ‖ PaperFlow: Profiling, Recommending, and Adapting Across Daily Paper Streams (PaperFlow)  ‖ OneChord: Distill What You Rank, Recommend in One Pass (OneChord) — arXiv 2610.25805|
| **University of Chinese Academy of Sciences (UCAS)** | LaRec: Unleashing LLM-based Latent Reasoning for Generative Recommendation (LaRec) ‖ CogRec: Structure-Cognitive Fast-and-Slow Reasoning for Generative Recommendation (CogRec) ‖ InforID: Adaptive Semantic Capacity Allocation for Parallel Generative Recommendation (InforID) ‖ D3ER: Supporting Multi-Modal Recommendation via Disentangle and Distillation-based Dynamic Ensemble (D3ER) ‖ Beyond Polarization: The Generative Constraint of Chain-of-Thought in Pointwise Reranking |
| **University of Electronic Science and Technology of China** | CFALR: Collaborative Filtering-Augmented Large Language Model for Personalized Fashion Outfit Recommendation (CFALR) ‖ CARD · ProMax · AdaSID · QuaSID · FAVE ‖ Adaptive Loss Balancing for Noise-Robust GRPO in Generative Recommendation (AdaGRPO) ‖ Measuring and Mitigating Identity-Cue Preference Drift in LLM-based Recommender Systems (PromptShift) ‖ Unleashing the Potential of Neighbors: Diffusion-based Latent Neighbor Generation for Session-based Recommendation (DiffSBR) — arXiv 2601.03903 ‖ RoleMix: Unifying Sequential and Non-Sequential Features via Semantic Tokenization for Post-Click Conversion Rate Prediction (RoleMix) — KDD Cup 2026 Tencent UniRec Challenge, arXiv 2607.22700 ‖ SPAR: Stepwise Personalized Autoregressive Recommendation with Prefix-Balanced Semantic IDs (SPAR) — arXiv 2610.27898 |
| **University of Georgia** | SAERec: Constructing Fine-grained Interpretable Intents Priors via Sparse Autoencoders for Recommendation (SAERec) ‖ Fair on the Surface? Benchmarking Hidden-Output Fairness Gaps in LLM Recommenders (FairGap) |
| **University of Glasgow** | RecRec: Latent Interests Recursive Reasoning for Sequential Recommendation (RecRec) ‖ Stream-aware Side Adaptation for Large Pre-trained Multimodal Embedding Models in Sequential Recommendation (Stresa) |
| **University of Gothenburg** | Designing Trustworthy LLM-based Wellbeing Recommendation through Controllable Interaction |
| **University of Hong Kong** | TimeRoute: Time-Aware Modality Routing and Diffusion for Multi-Modal Recommendation (TimeRoute) ‖ Agents as Knowledge Integrator and Utilizer in Multimodal Recommendation (AgentMMRec) ‖ OneLA: Scaling Linear-Attention Decoding to Large Beams in Generative Recommendation (OneLA) ‖ Beyond Successor Accuracy: State Retention for Recursive Self-Improvement in Recommendation (RecRSI) — arXiv 2610.07105 |
| **University of Illinois at Chicago** | Toward User Preference Alignment in LLM Recommendation via Explicit Context Feedback ‖ Scoring Is Not Enough: Addressing Gaps in Utility-fairness Trade-offs for Ranking ‖ Unbiased Recommender Systems with Implicit Feedback ‖ Rethinking Semantic ID Construction for Generative Recommendation: SimHash with Parallel Decoding and Semantic Alignment (FLASH) — arXiv 2610.07402 ‖ SafeCRS: Personalized Safety Alignment for LLM-Based Conversational Recommender Systems (SafeCRS) — arXiv 2603.03536 [opensource]|
| **University of Illinois Urbana-Champaign** | Personalized and Trust-Aware Health Recommendation Policies for a Construction Workplace (Trust-Aware Health Rec) |
| **University of Luxembourg** | Popcorn: A Configurable Benchmark for Visual Evidence in Multimodal Movie Recommendation (Popcorn) |
| **University of Maine at Presque Isle** | Diagnosing and Mitigating Retrieval Bottlenecks in LLM-Based Cold-Start Recommendation (LHF) |
| **University of Massachusetts Amherst** | Multimodal Music Recommendation System using LLMs ‖ Adaptive Re-Ranking |
| **University of Massachusetts Lowell** | TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation (TRACER) |
| **University of Melbourne** | Who You Are Adds Nothing Detectable to Where You Go Next: Sociodemographic Conditioning in LLM Next-Location Prediction (Next-Location) |
| **University of New South Wales** | How Reliable Are Semantic-ID Tokenizer Comparisons in Generative Recommendation? ‖ Beyond Item Order: Temporal Gap Tokenization for Generative Recommendation with Semantic IDs (ChronoSID) ‖ An Empirical Evaluation of Cross-City POI Recommendation on a Large-Scale Benchmark ‖ ChronicleRec: Pre-training Temporally Anchored Tokens for Lifelong User Modeling (ChronicleRec) ‖ Multimodal Graph Retrieval-Augmented Sequential Recommendation via Collaborative Filtering Paths (MGRASRec) — arXiv 2610.11228 |
| **University of Notre Dame** | Delegation Asymmetry in Agentic Recommender Systems: Measuring Two-Sided Receptivity in Online Dating — arXiv 2608.18058  ‖ AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned Prefixes (AlignPoison) — arXiv 2610.14487|
| **University of North Carolina at Chapel Hill** | ReasonRec: A Reasoning-Augmented Multimodal Agent for Unified Recommendation (ReasonRec) ‖ Diffusion-GR2: Diffusion Generative Reasoning Re-ranker (Diffusion-GR2) ‖ It Takes Two to Match: Co-Evolving Generative Retriever with Reinforcement Learning (CoGR)  ‖ AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned Prefixes (AlignPoison) — arXiv 2610.14487|
| **University of North Carolina at Charlotte** | Position: Recommender Systems Should Move Beyond Platform-Centric Ranking toward Personal Agent-Mediated Recommendation (PAMR) |
| **University of Otago** | How Reliable Are Semantic-ID Tokenizer Comparisons in Generative Recommendation? |
| **University of Ottawa** | Democracy Needs Reach: Political Equality, Online Speech, and Algorithmic Recommendation |
| **University of Pennsylvania** | LLM-Assisted Reranking to Operationalize Nuanced Objectives in Recommender Systems ‖ Safety as a Constraint: Fine-Tuning a LLM Recommender to Explain Itself ‖ Reinforcement Learning with Conformal Action Sets: An Application to Sequential Recommendation (RLCP) — arXiv 2610.08743 |
| **University of Pisa** | SURF: Subtractive Updates for Recommender Forgetting (SURF) |
| **University of Rochester** | ChronoID: Infusing Explicit Temporal Signals into Semantic IDs for Generative Recommendation (ChronoID) ‖ Decision-Oriented Recommendation Reranking: An Empirical Study of Jev (Jev) — arXiv 2609.40241 |
| **University of Science and Technology of China** | DualGR: Generative Retrieval with Long and Short-Term Interests Modeling (DualGR) ‖ QGS: Query-Conditioned Generative Search‖ DIRECTOR: Dynamic Index-based Recommendation with Transport-Optimized Retrieval (DIRECTOR) ‖ Think-to-Personalize: Unifying Reasoning and Retrieval for User-Centric Personalized Dense Retrieval (Think-to-Personalize) ‖ Native Multimodal Representation Learning for Click-Through Rate Prediction in E-Commerce Scenarios ‖ SWIM: Step-Wise Integrated Measure for Session-supervised List Evaluation in Generative Re-ranking (SWIM) ‖ VARG: Value-Aware and Ranking-Aligned Generative Retrieval for Dynamic E-commerce Search (VARG) ‖ FineSID: Scalable and Efficient Semantic Identifier Learning for Generative Recommendation — arXiv 2609.36670‖ Generative Recommenders Forget the Long Tail (CAPE) — arXiv 2610.29298 ‖ Knowledge Distillation across Item and Code Spaces for Parallel Generative Recommendation — arXiv 2610.32943 |
| **University of Science, VNU-HCM (Vietnam National University Ho Chi Minh City)** | rEDMRec: Distilling Large Language Model Reasoning into an Editable Experience Memory for Recommendation (rEDMRec) |
| **VinUniversity** | EPIC: Explicit Posterior Item Conditioning for Semantic ID Diffusion Recommendation (EPIC) |
| **University of Sheffield** | When and How to Ask: Dynamic Preference Elicitation Strategies for Conversational Recommendation (COPE) |
| **University of Siegen** | FINALLY: A Dataset Recommender System for Recommender-Systems Research (FINALLY) |
| **University of Stavanger** | A Standardized Re-evaluation of Conversational Recommender Systems on the ReDial Dataset |
| **University of Technology Sydney** | When Recommendation Denoising Meets Popularity Bias: Understanding and Mitigating Their Interaction (PAD) ‖ Segmentation-Supervised Complexity-Adaptive Recommendation (S2-CAR) ‖ TRUST: Item-Calibrated Interval Evidence for Temporal Session-Based Recommendation (TRUST) ‖ AgentSelect: Benchmark for Narrative Query-to-Agent Recommendation (AgentSelect) || Eval4DiRec: A Unified and Systematic Evaluation Framework for Diffusion-based Recommender Systems (Eval4DiRec) - ACM TKDD, arXiv 2609.34404 ‖ SPRINT: Single-Step Generative Recommendation via Average Probability Velocity (SPRINT) ‖ Generate What You Can Trust: Content Credibility in Generative Recommenders (CreGR)  ‖ RES-MR: Risk-Aware Reasoning for Explainable and Safe Medication Recommendation (RES-MR) — SIGIR 2026, arXiv 2610.14474|
| **University of Texas at Arlington** | Fairness Attacks on Recommender Systems |
| **University of Toronto** | Evaluating Scene-based In-Situ Item Labeling for Immersive Conversational Recommendation (ICRS) ‖ SafeGEO: Understanding Generative Engine Optimization Risks in Recommendation Agents (SafeGEO) ‖ ReFilter: Bridging Embeddings and LLM Filtering for Similar Mobile App Retrieval (ReFilter) — ASIS&T 2026, arXiv 2609.25306 |
| **University of Toronto / Microsoft Research** | From Noise to Order: Learning to Rank via Denoising Diffusion (DiffusionRank) |
| **Université du Québec à Montréal** | ReFilter: Bridging Embeddings and LLM Filtering for Similar Mobile App Retrieval (ReFilter) — ASIS&T 2026, arXiv 2609.25306 |
| **University of Southern California** | The Recall Ceiling of LLM Recommendation Reranking (RecallCeiling) — arXiv 2609.27953 ‖ Fair on the Surface? Benchmarking Hidden-Output Fairness Gaps in LLM Recommenders (FairGap) |
| **University of Tsukuba** | Generative Universal Multimodal Retrieval with Dual-role Identifiers (DrIG) |
| **University of Vermont** | BanglaShop-CRS: A User-Centric Bangla Dataset for Conversational Recommendation (BanglaShop-CRS) |
| **University of Victoria** | Gender and Race Bias in Consumer Product Recommendations by Large Language Models |
| **University of Virginia** | SAPO: Step-Aligned Policy Optimization for Reasoning-Based Generative Recommendation ‖ Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems (with Netflix, Cornell) ‖ Implicit Reasoning for Large Language Model-based Generative Recommendation (PauseRec) ‖ RecToolBench: Benchmarking Recommendation-Specific Tool Orchestration under Fuzzy User Intent (RecToolBench) |
| **University of Washington** | Cold-Start Personalization via Training-Free Priors from Structured World Models (PEP) |
| **University of Waterloo** | CompRank: Efficient LLM Reranking via Token-Level Compression and Decoding-Free Scoring (CompRank) ‖ Do Recommendation Algorithms Work When Users Are LLM Agents? A Case Study on Moltbook (Moltbook) |
| **University of Wisconsin-Madison** | Grounded Token Initialization for New Vocabulary in LMs for Generative Recommendation (GTI) |
| **University of Zurich** | Attention Calibration for Position-Fair Dense Information Retrieval ‖ The Utility of LLMs in Recommender Systems Explanation Evaluation ‖ Reproducing Transparent and Scrutable Recommendations: Exploring Open-Weight Models via Natural-Language User Profiles (Transparent UPR Repro) |
| **Universität Augsburg** | Guiding Federated Graph Recommendation with LLM-encoded knowledge |
| **Université de Montréal** | Building a privacy-preserving Federated Recommender system for mobile devices || Mult-BiW: Mitigating Popularity Bias in Recommendation (Mult-BiW) - ACM TOIS, arXiv 2609.35041 |
| **Unknown (Academic)** | TriAlignGR: Triangular Multitask Alignment with Multimodal Deep Interest Mining for Generative Recommendation (TriAlignGR) |
| **Unknown (Industrial)** | CapsID: Soft-Routed Variable-Length Semantic IDs for Generative Recommendation (CapsID) |
| **USTC** | The Pitfall of Scaling Up: Uncovering and Mitigating Popularity Bias Amplification in Scaling Transformer-based Recommenders (SPRINT) ‖ MiniOneRec · UniGRec · PAD-Rec · DynamicPO: Dynamic Preference Optimization for Recommendation · Objective Shaping · Beyond Static Best-of-N: Bayesian List-wise Alignment for LLM-based Recommendation (BLADE) ‖ QGS: Query-Conditioned Generative Search ‖ Intuition-Guided Latent Reasoning for LLM-Based Recommendation (IntuRec) ‖ RSIR: Can Recommender Systems Teach Themselves? (RSIR) ‖ Mitigating Reward Hacking in LLM-based Recommendation (SIRIUS) ‖ LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation (LBR) ‖ Autonomous Information Seeking: A Roadmap for Agentic Recommender Systems ‖ Seeing and Reflecting: Multimodal Memory-Enhanced Agent Collaboration for Recommendation (MMEACR) ‖ Beyond Noisy Signals: Dual-Level Denoising for Multi-modal Sequential Recommendation (DDMSR) ‖ IMFuse: Instance-Aware Multi-Layer Fusion for LLM-Enhanced Sequential Recommendation (IMFuse) ‖ Uncertainty-aware Generative Recommendation (UGR) ‖ SITA: Semantic Interest Tokens for Target-Aware Compression in Long-Sequence Recommendation (SITA) ‖ Rethinking Item Tokenization in Generative Recommenders: From Fixed Atoms to Semantic Subwords (SST) ‖ Preference Shapes Relevance: Cross-component Hierarchical Semantic Alignment for Personalized Generative Retrieval (CHAP) |
| **UT Austin** | RRCM ‖ LLM Agents Enable User-Governed Personalization Beyond Platform Boundaries |
| **Vienna University of Economics and Business** | Understanding and Debugging Failures in N-Gram-Based Generative Retrieval |
| **Virginia Tech** | Bi-NAS: Towards Effective and Personalized Explanation for Recommender Systems via Bi-Level Neural Architecture Search (Bi-NAS) ‖ HypRQ-VAE: Hyperbolic Item Indexing for Long-Tail-Aware Generative Recommender Systems (HypRQ-VAE) |
| **Vrbo / Expedia Group** | Serving the Long Tail: Training-Free LLM Candidate Generation for Vacation Rental Marketplaces |
| **VWFS** | VRec · Rethinking Convolutional Networks for Attribute-Aware Sequential Recommendation (ConvRec) |
| **Walmart Global Tech** | CRAB ‖ INSPIRE: Intent-aware Neural Sponsored Product Retrieval for E-commerce ‖ Scaling Dense Retrieval with LLM-Annotated Training Data: Structured Mining and Progressive Curriculum for E-Commerce Sponsored Search ‖ Unified Multi-Task Relevance Modeling for E-Commerce: Comparing Task Routing Architectures Across LLMs and Cross-Encoders ‖ LLM-HYPER: Generative CTR Modeling for Cold-Start Ad Personalization via LLM-Based Hypernetworks (LLM-HYPER) ‖ Beyond Ranking Accuracy: Evaluating LLM-Cited Feature Rationales for Next Basket Repurchase Recommendation ‖ RegionFed: Federated Learning for Personalized Query Understanding in Heterogeneous Retail Environments (RegionFed) |
| **Waseda University** | Adaptive Loss Balancing for Noise-Robust GRPO in Generative Recommendation (AdaGRPO) || ED-DR: Off-Policy Ranking Evaluation under Ranking-Dependent Examination (ED-DR) - arXiv 2609.35034 |
| **Washington University in St. Louis** | PAPA: Online Personalized Active Preference Alignment (PAPA) |
| **Webtoon (NAVER WEBTOON)** | Progressive Alignment of Recommender Foundation Model through Multi-Phase Post-Training (Progressive FM Post-Training) |
| **Wolt (DoorDash, Inc.)** | Balancing Trial and Reorder: A Hybrid Sequential Transformer-GBDT Ranker for On-Demand Delivery (UVR) |
| **Wuhan University** | GRC: Learning to Reflect and Correct (GRC) ‖ Think2Go: Generative Next POI Recommendation with LLM Reasoning (Think2Go) ‖ MTServe ‖ Tail-Aware Adaptive-k: Query-Adaptive Context Selection for Retrieval-Augmented Generation (TAA-k) ‖ MDGR: A Masked Diffusion Generative Recommendation Framework (MDGR) ‖ Reasoning Quality Matters: Combating Reasoning Collapse in LLM-based Embedding Learning (CoFree) |
| **Xi'an Jiaotong University** | Harmonizing Semantic and Collaborative in LLMs: Reasoning-based Embedding Generator for Sequential Recommendation (ReaEmb) ‖ The Best of Both Worlds: Harmonizing Semantic and Hash IDs for Sequential Recommendation (H²Rec) |
| **Xi'an Jiaotong-Liverpool University** | HiLaR: Hierarchical Latent Reasoning for LLM-based Recommendation (HiLaR) ‖ AtomRec: Evolving Atomic Memory for Agentic Recommendation (AtomRec) |
| **Xiamen University** | ManCAR: Manifold-Constrained Latent Reasoning with Adaptive Test-Time Computation for Sequential Recommendation (ManCAR) |
| **Xiaohongshu** | LIFT: Lifecycle-Aware Interaction Factorization for Unified Retrieval and Ranking (LIFT) - arXiv 2610.10556 ‖ Vela: User Behavior as Supervision Sparks LLM Intelligence (Vela) — AllSpark at Xiaohongshu (RedNote), arXiv 2610.14709 [opensource] |
| **Xidian University** | AgentGR: Semantic-aware Agentic Group Decision-Making Simulator for Group Recommendation ‖ EviRank: Evidence-Based Confidence Estimation for LLM-Based Ranking (EviRank) ‖ ANCHOR: Agentic Noise Creation Framework for Human Simulation and Denoising Recommendation (ANCHOR) ‖ Differentiable Geometric Indexing for End-to-End Generative Retrieval (DGI) ‖ TopoGR: Revealing and Preserving Latent Structure of Semantic ID in Generative Recommendation (TopoGR) |
| **Yale University** | CoPersona: Collaborative Persona Graphs for Robust LLM Personalization (CoPersona) ‖ Massive Memorization with Hundreds of Trillions of Parameters (VISTA) |
| **Yandex** | Gated Bidirectional Linear Attention for Generative Retrieval (GBLA) ‖ Gryphon: A Unified Architecture for Semantic-ID Generation and Item-Level Scoring in Industrial Recommendations (Gryphon) ‖ Gryphon-v2: One Model in Place of a Cascade — Generate-and-Rank Recommender with Rollout Distillation (Gryphon-v2) ‖ Long-History User Transformers for Real-Time Ad Ranking ‖ Sona Technical Report (Sona) ‖ A Redundancy Reduction Approach for Controllable Sequential Recommendations (BT-SR) |
| **Yonsei University** | SELF-INDEX: Self-Evolving Search Index (SELF-INDEX) — with Samsung Research, UC Irvine & Korea University ‖ Reasoning with Evidence, Not Merely Rationales: Verifiable Preference Proofs for LLM-Based Recommendation (PROVE-REC) — arXiv 2610.02968 ‖ MATE: Adaptive Long- and Short-Term User Memory for LLM-Based Recommendation (MATE) — arXiv 2610.06050|
| **Youtube** | [STATIC] Vectorizing the Trie |
| **Z Lab (Chengdu, Sichuan)** | Dense Feature Representation over Sequence Modeling: A Solution to the KDD Cup 2026 UniRec Challenge (KDD Cup 2026 UniRec) |
| **Zalando** |  VCG: A Multimodal Retrieval Framework for E-Commerce Video Feeds under Extreme Cold-Start Conditions (VCG) ‖ On Evaluating and Improving Conversational Agents in Production — arXiv 2609.32092 |
| **ZBot Technology** | SafeGEO: Understanding Generative Engine Optimization Risks in Recommendation Agents (SafeGEO) |
| **Zhejiang University** | Think When Needed: Model-Aware Reasoning Routing for LLM-based Ranking ‖ The Pitfall of Scaling Up: Uncovering and Mitigating Popularity Bias Amplification in Scaling Transformer-based Recommenders (SPRINT) ‖ Semantic Trimming and Auxiliary Multi-step Prediction for Generative Recommendation (STAMP) ‖ DeGRe: Dense-supervised Generative Reranking for Recommendation ‖ TAP-PER: Beyond Retrieval: Learning Compact User Representations for Scalable LLM Personalization (TAP-PER) ‖ OneBar: An End-to-End Content-Grounded Generative Query Recommendation Framework for E-Commerce Video Feeds (OneBar) ‖ LBR: Towards Mitigating Length Bias in Large Language Models for Recommendation (LBR) ‖ FORGE: Forming Semantic Identifiers for Generative Retrieval in Industrial Datasets (FORGE) ‖ TSGR: Taobao Search Generative Retrieval (TSGR) ‖ IMFuse: Instance-Aware Multi-Layer Fusion for LLM-Enhanced Sequential Recommendation (IMFuse) ‖ SmartGR: Hierarchy and Beam-Aware Knowledge Distillation for Generative Recommendation (SmartGR) ‖ Making Collaborative Signals Count: Graph-Aware Large Language Models for Sequential Recommendation (GALLM) ‖ Astar: Learning to Propose Evolution Directions for Self-Evolving Industrial AI Systems (Astar) ‖ World Model-Guided Reinforcement Learning via Counterfactual User Engagement Simulation (WMG-RL) ‖ RegRet: Enhancing Region-Level Retrieval in Large Multimodal Models (RegRet) — ECCV 2026 ‖ Learning Robust Personalized Prompts for LLM-Driven Sequential Recommendation (LRPRec) ‖ Training with Missed Targets in Generative Recommendation (Missed Targets) — arXiv 2610.10124 ‖ SharpRec: Sharpness-aware Model Merging with Salience Recovery for LLM-based Cross-Domain Sequential Recommendation (SharpRec) — KDD 2026, arXiv 2607.25366|
| **Zhengzhou University** | IMFuse: Instance-Aware Multi-Layer Fusion for LLM-Enhanced Sequential Recommendation (IMFuse) |
| **Yu Sun (affiliation not disclosed on arXiv)** | Beyond Semantic IDs: Finite-Beam Reachability and Continuous-Query Generation in Generative Recommendation (Beyond Semantic IDs) -- arXiv 2610.12977 |
| **AIRI (Artificial Intelligence Research Institute)** | Separating Thinking from Spelling for Efficient Generative Recommendation (Deco) — arXiv 2610.32980 |
| **Skolkovo Institute of Science and Technology** | Separating Thinking from Spelling for Efficient Generative Recommendation (Deco) — arXiv 2610.32980 |
| **University of Massachusetts Boston** | FAIRGate: Agentic Runtime Policy Selection and Verified Repair for Fairer Recommendation Delivery (FAIRGate) — arXiv 2610.30359 |