RL from Human Feedback

RL: Reinforcement Learning

Overview

RLHF studies how to use human judgments and interventions to teach AI systems which responses or actions are desirable, especially when goals are hard to specify with a fixed reward. Common approaches learn reward models from pairwise comparisons and optimize policies with PPO or GRPO, while methods such as Direct Preference Optimization learn directly from preference pairs; active learning and online feedback aim to reduce the data required. Research spans language-model alignment and benchmarks such as AlpacaEval and MT-Bench, as well as robot learning from corrections and demonstrations. A central challenge is that feedback can be noisy, biased, or diverse, and optimizing an imperfect reward can cause reward hacking, so work increasingly studies uncertainty, robustness, fairness, and how to measure real-world improvement rather than proxy scores alone.

Latest papers

CardsList

136 papers9 papers in the last four weeks, down 10% on the four weeks before. 0.1% of all new papers.

  1. CRAFT: A Robot-Agent Harness with Scope-Aware Adaptation

    Nov 15, 2026Peng Zhang, Jiawen Wang, Cheng Xue +1Robot Policy AdaptationMemory-Augmented Agents

  2. Leveraging Psychophysical Attentional Distribution for Gaze-Augmented Reward Modeling

    Nov 10, 2026Chaeho Lee, Junyup Kim, Wansoo Kim +2Reward ModelingPreference Learning for LLMs

  3. AERO: Adaptive Ensemble-Disagreement Routing for Oracle Feedback in Sample-Efficient Online RLHF

    Nov 7, 2026Shrey Singhal, Nannan Huang, Haytham M. FayekLLM AlignmentRL from Human Feedback

  4. Q-OIL: Optimistic Exploration in Learning from Human Interventions

    Nov 5, 2026Pranav Teegavarapu, Sriyash Poddar, Rico Qi +5RL for RoboticsHuman-in-the-Loop AI

  5. Provable Zeroth-Order Safe Exploration for Reinforcement Learning from Human Feedback

    Nov 5, 2026Weiqin Chen, Xinmeng Huang, Dongsheng Ding +1Safe RLReinforcement Learning

  6. Environment Edits as a Corrective Signal for Reward Learning

    Oct 26, 2026Isaac Ngui, Courtney McBeth, Nancy M. AmatoRobot LearningPreference Learning

  7. Human-Feedback-Guided Multi-Agent Inverse Reinforcement Learning for Interactive Driving Behavior Modeling

    Oct 24, 2026Longhao Yan, Jinhao Liang, Kaidi YangReinforcement LearningRL for Autonomous Driving

  8. Aligning Large Language Models with Human Feedback: Mathematical Foundations and Algorithm Design

    Oct 22, 2026Siliang Zeng, Luca Viano, Chenliang Li +6LLM AlignmentRL from Human Feedback

  9. Differentially Private Sparse Reward Estimation with Preference Feedback

    Oct 21, 2026Meng Ding, Mingxi Lei, Jie Zhang +2Reward ModelingRL from Human Feedback

  10. Reason-guided Human Reward Modeling for Preference-based Reinforcement Learning

    Oct 16, 2026Anshul Nayak, Shahil Shaik, Jonathan Smereka +1Preference-Based RLReward Modeling

  11. CleanRM: Reward Modeling from Noisy Feedback with a Few Clean Feedback for RLHF

    Oct 15, 2026Licheng Pan, Haocheng Yang, Yuzhou Chen +7RL from Human FeedbackNoisy-Label Learning

  12. A Scoping Review and Experimental Study on Reinforcement Learning from Human Feedback for Human-Robot Collaboration

    Oct 7, 2026Alexandra Coroiu, Andrea Vogt, Viktor Werbilo +3RL from Human FeedbackHuman-Robot Collaboration

  13. Constraint Tree Exploration for Learning from Language Feedback

    Oct 6, 2026Shaoang Li, Daniel R. Jiang, Jian LiError FeedbackRL Exploration

  14. EnGRICH: Enhancing Generative Reward Modeling with Critiques from Humans

    Oct 4, 2026Xuancheng Li, Beining Wang, Haitao Li +7Reward ModelingProcess Reward Models

  15. Robust Risk-Sensitive Reinforcement Learning from Corrupted Human Feedback

    Sep 30, 2026Xinyi Ni, Lifeng LaiRisk-Sensitive RLRobust RL

  16. ReF-HIL: Shaping the Critic around Human Action Neighborhoods for Efficient Human-in-the-Loop Reinforcement Learning

    Sep 29, 2026Shaoyin Luo, Song Wang, Shibo Xia +5RL for RoboticsReward Shaping

  17. Diffusion Reward Models

    Sep 27, 2026Xiangyang Wang, Bingxiang He, Zeyuan Liu +13Reward ModelingMultimodal Reward Modeling

  18. RMB: Reward Model Boosting Mitigates Reward Hacking

    Sep 27, 2026Jiabin Fan, Dezhi Ye, Yongchang Hao +1Reward HackingRL from Human Feedback

  19. BEE: Intervention-Adaptive Real-World Reinforcement Learning with Vision-Language-Action Models

    Sep 23, 2026Weihui Zhao, Xiaohan Yan, Zunian Wan +14RL for VLMsRobotic Manipulation

  20. Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Aligner

    Sep 11, 2026Kazusato Oko, Annie Ulichney, Nika Haghtalab +1LLM AlignmentPreference Alignment

  21. Subspace Inference Enables Efficient Active Reward Learning from Preferences

    Sep 3, 2026Yutai Zhou, Erdem BıyıkBayesian RLActive Preference Learning

  22. User Feedback Provides a Unique Signal that LLMs Can not Detect

    Sep 2, 2026Shachar Don-Yehiya, Leshem Choshen, Omri AbendLLM EvaluationRL from Human Feedback

  23. From Answers to Policies: Efficient In-Context Learning System through Emulating Expert Investigation

    Aug 17, 2026Minh-Ha Nguyen, Ngoc-Ngo Quang Tran, Thuy Dung Nguyen +1In-Context LearningIn-Context RL

  24. The Ethical Decision Head: Operationalizing Normative Ethics in Autonomous Vehicles via Reinforcement Learning from Human Feedback

    Aug 17, 2026Thomas Mbrice, Ammar Ali, Sami Mian +5RL from Human FeedbackRL for Autonomous Driving

  25. GAINS: Leveraging Inconsistent Human Intervention Signals in Reinforcement Learning

    Aug 16, 2026Xinyi Zhang, Yinuo Zhao, Pei Ren +7Robotic RLRL from Human Feedback

  26. Why Summaries Turn Neutral: Policy Attribution for Sentiment Drift in Reinforcement Learning from Human Feedback

    Aug 16, 2026Mikhail Krasitskii, Alexander Gelbukh, Olga Kolesnikova +1RL from Human FeedbackText Summarization

  27. Learning from Online User Feedback for Shopping Agents

    Aug 12, 2026Haobo Zhang, Kelong Mao, Sulong Xu +2Large Language Model-Based RecommendationUser Preference Modeling

  28. Procedural Fairness Failures in RLHF from Preference Averaging

    Aug 10, 2026M P V S Gopinadh, Karthik Kamuju, Kummari Avinash +2Reward ModelingAlgorithmic Fairness

  29. LEMUR: Learning to Align with Multi-Objective Reinforcement Learning from Preference Feedback

    Jul 31, 2026Manith Adikari, Bei Peng, Samuele Vinanzi +1Multi-Objective Reinforcement LearningReinforcement Learning