Reinforcement Learning from Human Feedback

Automated Filtering of Human Feedback Data for Aligning Text-to-Image Diffusion Models ICLR 2025 BLCU-ICALL at BEA 2025 Shared Task: Multi-Strategy Evaluation of AI Tutors ACL 2025 CAPO: Confidence Aware Preference Optimization Learning for Multilingual Preferences IJCNLP 2025 CARE: Multilingual Human Preference Learning for Cultural Awareness EMNLP 2025 CARMO: Dynamic Criteria Generation for Context Aware Reward Modelling ACL 2025 Cheems: A Practical Guidance for Building and Evaluating Chinese Reward Models from Scratch ACL 2025 Common Learning Constraints Alter Interpretations of Direct Preference Optimization AISTATS 2025 Comparing Bad Apples to Good Oranges Aligning Large Language Models via Joint Preference Optimization ACL 2025 Correlated Proxies: A New Definition and Improved Mitigation for Reward Hacking ICLR 2025 Corruption Robust Offline Reinforcement Learning with Human Feedback AISTATS 2025 COS-DPO: Conditioned One-Shot Multi-Objective Fine-Tuning Framework UAI 2025 Debiasing Online Preference Learning via Preference Feature Preservation ACL 2025 Detecting and Mitigating Hallucination in Large Vision Language Models via Fine-Grained AI Feedback AAAI 2025 Direct Judgement Preference Optimization EMNLP 2025 DISCO Balances the Scales: Adaptive Domain- and Difficulty-Aware Reinforcement Learning on Imbalanced Data EMNLP 2025 DUO: Diverse, Uncertain, On-Policy Query Generation and Selection for Reinforcement Learning from Human Feedback AAAI 2025 Enhancing RLHF with Human Gaze Modeling EMNLP 2025 Exposing Privacy Gaps: Membership Inference Attack on Preference Data for LLM Alignment AISTATS 2025 Finding the Sweet Spot: Preference Data Construction for Scaling Preference Optimization ACL 2025 From Outcomes to Processes: Guiding PRM Learning from ORM for Inference-Time Alignment ACL 2025 Generative Reward Modeling via Synthetic Criteria Preference Learning ACL 2025 Henry at BEA 2025 Shared Task: Improving AI Tutor’s Guidance Evaluation Through Context-Aware Distillation ACL 2025 How to Mitigate Overfitting in Weak-to-strong Generalization? ACL 2025 Hybrid Preferences: Learning to Route Instances for Human vs. AI Feedback ACL 2025 Improve LLM-as-a-Judge Ability as a General Ability EMNLP 2025