Beyond Static Best-of-N: Bayesian List-wise Alignment for LLM-based Recommendation
sigir26-011 · Full research paper · Ruijun Chen, Chongming Gao, Jiawei Chen, Weiqin Yang, Xiangnan He
Knowledge Infusion and Open-World Au · 50% Preference Alignment and Optimizatio · 50%
Abstract
Large Language Models have revolutionized recommender systems (LLM4Rec) by leveraging their generative capabilities to model complex user preferences. However, existing LLM4Rec methods primarily rely on token-level objectives, making it difficult to optimize list-level and non-differentiable metrics (e.g., NDCG, fairness) that define actual recommendation quality. While Best-of-N (BoN) directly optimizes these metrics during inference, its high computational cost hinders real-world deployment. To address this, BoN Alignment aims to distill the search capability into the model itself, yet current approaches suffer from two critical limitations: (1) Indiscriminate Supervision, where the static reference fails to distinguish the relative quality of candidates exceeding its empirical range, leading to a loss of ranking guidance; and (2) Gradient Decay, where the effective supervision signal rapidly diminishes as the evolving policy improves, resulting in inefficient optimization.
Assigned reviewers
| # | Reviewer | Match | Why (closest evidence publications in the taxonomy) | |
|---|---|---|---|---|
| 1 | Xiao Zhang 0034 | 47 | 78 Large Language Models Enhanced Collaborative Filtering (2024) Preference Alignment and Optimization 54 Similarity = Value? Consultation Value-Assessment and Alignment for Personalized Search (2025) Personalized and Context-Aware Retrieval 51 UniSAR: Modeling User Transition Behaviors between Search and Recommendation (2024) Recommendation with Search and Ranking |
|
| 2 | Min Hou 0004 | 47 | 75 RecCocktail: A Generalizable and Efficient Framework for LLM-Based Recommendation (2026) Preference Alignment and Optimization 55 WeaveRec: An LLM-Based Cross-Domain Sequential Recommendation Framework with Model Merging (2026) Efficiency and Deployment 48 Mitigating Distribution Shifts in Sequential Recommendation: An Invariance Perspective (2025) Contrastive and Self-Supervised Sequential Recommendation |
Area chair
| Area chair | Match | |
|---|---|---|
| Pengyue Jia | 47 |
Area-chair candidates (6)
| Pengyue Jia assigned | 47 | 10 papers | |
| Leyu Lin | 46 | 15 papers | |
| Yanan Niu | 45 | 15 papers | |
| Kun Gai | 45 | 5 papers | |
| Maolin Wang 0001 | 45 | 15 papers | |
| Xin Xin 0003 COI: co-author | 44 | 10 papers |
Candidates from the panel ranked by match score
| # | Reviewer | Match | Load | Why | |
|---|---|---|---|---|---|
| 1 | Zhengyi Yang 0007 COI: co-author | 66 | 3 | 73 On Softmax Direct Preference Optimization for Recommendation (2024) Preference Alignment and Optimization 73 AlphaDPO: Adaptive Reward Margin for Direct Preference Optimization (2025) Preference Alignment and Optimization |
|
| 2 | Chongming Gao COI: author | 58 | 3 | 75 DLCRec: A Novel Approach for Managing Diversity in LLM-Based Recommender Systems (2025) Preference Alignment and Optimization 75 MGFRec: Towards Reinforced Reasoning Recommendation with Multiple Groundings and Feedback (2026) Preference Alignment and Optimization |
|
| 3 | Jiancan Wu COI: co-author | 56 | 3 | 73 Larger or Smaller Reward Margins to Select Preferences for LLM Alignment? (2025) Preference Alignment and Optimization 73 AlphaDPO: Adaptive Reward Margin for Direct Preference Optimization (2025) Preference Alignment and Optimization |
|
| 4 | Bo Chen 0023 COI: co-author | 55 | 3 | 80 Large language models make sample-efficient recommender systems (2025) Efficiency and Deployment 80 How Can Recommender Systems Benefit from Large Language Models: A Survey (2025) Efficiency and Deployment |
|
| 5 | Jizhi Zhang COI: co-author | 55 | 3 | 73 TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation (2023) Preference Alignment and Optimization 73 Large Language Models are Learnable Planners for Long-Term Recommendation (2024) Preference Alignment and Optimization |
|
| 6 | Yang Zhang 0072 COI: co-author | 53 | 3 | 73 Reason4Rec: Deliberative User Preference Alignment of Large Language Models for Recommendation (2026) Preference Alignment and Optimization 73 TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation (2023) Preference Alignment and Optimization |
|
| 7 | Jieming Zhu | 52 | 3 | 78 Efficient and Deployable Knowledge Infusion for Open-World Recommendations via Large Language Models (2026) Knowledge Infusion and Open-World Augmentation 77 Towards Open-World Recommendation with Knowledge Augmentation from Large Language Models (2024) Knowledge Infusion and Open-World Augmentation |
|
| 8 | Shuchang Liu 0001 COI: co-author | 49 | 3 | 76 DLCRec: A Novel Approach for Managing Diversity in LLM-Based Recommender Systems (2025) Preference Alignment and Optimization 76 Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5) (2022) Knowledge Infusion and Open-World Augmentation |
|
| 9 | Bowen Zheng 0005 | 49 | 3 | 58 Autonomous Reasoning-Retrieval for Large Language Model Based Recommendation (2025) Reasoning-Intensive Retrieval and Ranking 53 Enhancing Sequential Recommender with Large Language Models for Joint Video and Comment Recommendation (2025) Sequential and Session-based Recommendation |
|
| 10 | Hao Liu 0078 | 48 | 3 | 55 WeaveRec: An LLM-Based Cross-Domain Sequential Recommendation Framework with Model Merging (2026) Efficiency and Deployment 48 Mitigating Distribution Shifts in Sequential Recommendation: An Invariance Perspective (2025) Contrastive and Self-Supervised Sequential Recommendation |
|
| 11 | Jianling Wang | 48 | 3 | 66 Fresh Content Recommendation at Scale: A Multi-funnel Solution and the Potential of LLMs (2024) Efficiency and Deployment 54 LLMs for User Interest Exploration in Large-scale Recommendation Systems (2024) Sequential and Session-based Recommendation |
|
| 12 | Xuanhua Yang | 48 | 3 | 69 Breaking the Relevance-Diversity Seesaw: Hierarchical LLM Reasoning with RL for Industrial Novelty Recommendation (2026) Preference Alignment and Optimization 56 GenRec: A Preference-Oriented Generative Framework for Large-Scale Recommendation (2026) Generative Retrieval and Ranking |
|
| 13 | Xiao Zhang 0034 assigned | 47 | 3 | 78 Large Language Models Enhanced Collaborative Filtering (2024) Preference Alignment and Optimization 54 Similarity = Value? Consultation Value-Assessment and Alignment for Personalized Search (2025) Personalized and Context-Aware Retrieval |
|
| 15 | Min Hou 0004 assigned | 47 | 3 | 75 RecCocktail: A Generalizable and Efficient Framework for LLM-Based Recommendation (2026) Preference Alignment and Optimization 55 WeaveRec: An LLM-Based Cross-Domain Sequential Recommendation Framework with Model Merging (2026) Efficiency and Deployment |
|
| 16 | Yupeng Hou | 46 | 3 | 68 Large Language Models are Zero-Shot Rankers for Recommender Systems (2024) Efficiency and Deployment 53 Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation (2024) Collaborative-Semantic Representation Alignment |
|
| 17 | Qijiong Liu | 46 | 3 | 70 RecBase: Generative Foundation Model Pretraining for Zero-Shot Recommendation (2025) Generative Retrieval and Recommendation Architectures 64 ONCE: Boosting Content-based Recommendation with Both Open- and Closed-source Large Language Models (2024) LLM-Derived Knowledge Infusion |
|
| 18 | Bo Peng 0009 | 46 | 3 | 55 eCeLLM: Generalizing Large Language Models for E-commerce from Large-scale, High-quality Instruction Data (2024) Efficiency and Deployment 50 M2: Mixed Models With Preferences, Popularities and Transitions for Next-Basket Recommendation (2023) Efficient and Long-Sequence Sequential Recommendation |
|
| 19 | Junjie Zhang 0009 | 46 | 3 | 56 STARec: An Efficient Agent Framework for Recommender Systems via Autonomous Deliberate Reasoning (2025) RL Training of Search Agents 55 Tapping the Potential of Large Language Models as Recommender Systems: A Comprehensive Framework and Empirical Analysis (2025) Efficiency and Deployment |
|
| 20 | Huifeng Guo COI: co-author | 46 | 3 | 55 SELF: Surrogate-light Feature Selection with Large Language Models in Deep Recommender Systems (2025) LLM-Derived Knowledge Infusion 55 LLMTreeRec: Unleashing the Power of Large Language Models for Cold-Start Recommendations (2025) Generative Recommendation and Item Tokenization |
|
| 22 | Keqin Bao COI: co-author | 46 | 3 | 73 TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendation (2023) Preference Alignment and Optimization 69 Item-side Fairness of Large Language Model-based Recommendation System (2024) Preference Alignment and Optimization |
|
| 23 | Lantao Hu | 45 | 3 | 51 Comprehensive List Generation for Multi-Generator Reranking (2025) Recommendation with Search and Ranking 51 Unsupervised Ranking Ensemble Model for Recommendation (2024) Recommendation with Search and Ranking |
|
| 24 | Seongmin Park 0002 | 45 | 3 | 49 Temporal Linear Item-Item Model for Sequential Recommendation (2025) Efficient and Long-Sequence Sequential Recommendation 48 MergeRec: Model Merging for Data-Isolated Cross-Domain Sequential Recommendation (2026) Cross-Domain and Cross-Market Sequential Recommendation |
|
| 25 | Wei Guo 0006 COI: co-author | 45 | 3 | 67 Can LLMs Outshine Conventional Recommenders? A Comparative Evaluation (2025) Efficiency and Deployment 48 User Behavior Modeling with Deep Learning for Recommendation: Recent Advances (2023) Recommendation Systems and Evaluation |
|
| 27 | Fei Sun 0001 COI: co-author | 45 | 3 | 51 Personalized re-ranking for recommendation (2019) Recommendation with Search and Ranking 50 Neural Re-ranking in Multi-stage Recommender Systems: A Review (2022) Recommendation Systems and Evaluation |
|
| 28 | Junwei Pan | 45 | 3 | 58 Cross-Scale Collaboration between LLMs and Lightweight Sequential Recommenders with Domain-Specific Latent Reasoning (2026) Generative Recommendation and Item Tokenization 53 Empowering Large Language Model for Sequential Recommendation via Multimodal Embeddings and Semantic IDs (2025) Generative Recommendation and Item Tokenization |
|
| 29 | Yimeng Bai | 45 | 3 | 48 LabelCraft: Empowering Short Video Recommendations with Automated Label Crafting (2024) Recommendation Systems and Evaluation 48 Leveraging Watch-time Feedback for Short-Video Recommendations: A Causal Labeling Framework (2023) Recommendation Systems and Evaluation |
|
| 32 | Han Li 0005 | 45 | 3 | 53 Enhancing Sequential Recommender with Large Language Models for Joint Video and Comment Recommendation (2025) Sequential and Session-based Recommendation 51 Comprehensive List Generation for Multi-Generator Reranking (2025) Recommendation with Search and Ranking |
|
| 33 | Ed H. Chi | 45 | 3 | 50 PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations (2026) Item Tokenization and Semantic IDs 48 DSelect-k: Differentiable Selection in the Mixture of Experts with Applications to Multi-Task Learning (2021) Multi-Task and Multi-Objective Optimization |
|
| 34 | Lichan Hong | 45 | 3 | 50 Improving Training Stability for Multitask Ranking Models in Recommender Systems (2023) Recommendation Systems and Evaluation 49 PLUM: Adapting Pre-trained Language Models for Industrial-scale Generative Recommendations (2026) Generative Recommendation and Item Tokenization |
|
| 35 | Jiangxia Cao | 45 | 3 | 54 Enhancing Content-based Recommendation via Large Language Model (2024) Collaborative-Semantic Representation Alignment 48 Contrastive Cross-Domain Sequential Recommendation (2022) Cross-Domain and Cross-Market Sequential Recommendation |