method · first seen 4 Oct 2026
reinforcement learning from human feedback
Also: RLHF, preference-based post-training, learning from human preferences, reward modeling from human feedback
Post-training a model with a reward model learned from human preference or quality judgments, here applied to video prediction.
Reinforcement learning from human feedback trains a reward model on human preference or quality judgments, then optimizes a policy against it. The sources treat it as standard background rather than a new idea: papers on multi-objective alignment, social-choice distortion guarantees, sycophancy origins, debate and cognitive bias, decentralized preference data, and active querying for online RLHF all cite it as the baseline. Extensions appear in a robotics paper on reward hacking by sampling-based planners, a Tencent Hunyuan repo for emotional video generation rewards, and flow-model reward post-training. A Show HN post offers tutorial notes. None of this introduces RLHF itself.
RLHF has been central to LLM alignment since at least 2022, so the rise reflects ordinary research volume rather than a new concept. The more informative signals are its spread to video, audio and robotics, and the continuing work on its failure modes such as reward hacking and sycophancy.
Analyst note (established), 17 Nov 2026.
Mentions per week
PapersCodeDiscussion
Rank 33 on 14 Nov 2026 · 17 mentions in 28 days · 17 independent groups · papers → code → discussion · first seen 6 weeks ago · acceleration +1.30
Earliest sources
- Which Preferences to Train On? End-to-End Multi-Objective Alignment with an Adversarial Preference DistributionPapers · 4 Oct 2026 · Minjae Lee, Kyunghyun Cho, Sangdon Park
- Groupwise Distortion Guarantees for Preference-Based AlignmentPapers · 4 Oct 2026 · Jacob Brodkey, Roberto Tamez, Aaron Roth
- AI Safety via Debate is Compromised by Cognitive BiasesPapers · 4 Oct 2026 · Gefei Liu, Sonya Rashkovan, Sophia Lloyd George, Isaac Sheidlower, Serena Booth
- MEND: RL For Flow Models via Proximal Velocity MatchingPapers · 5 Oct 2026 · Shreshth Saini, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C. Bovik
- A Scoping Review and Experimental Study on Reinforcement Learning from Human Feedback for Human-Robot CollaborationPapers · 7 Oct 2026 · Alexandra Coroiu, Andrea Vogt, Viktor Werbilo, Andreas Poppele, Johann Christens…
- I would rather quit NLP than read another paper like this: The rise of antithesis in NLP papersPapers · 7 Oct 2026 · Olga Zamaraeva, Adrián Gude, Roi Santos-Ríos, Carlos Gómez-Rodríguez
Latest sources
- A Theory of Policy Optimization in Post-Training: Coverage Collapse and Reference ResetsPapers · 10 Nov 2026 · Audrey Huang, Nan Jiang, Sham M. Kakade, Sadhika Malladi, Akshay Krishnamurthy,…
- Enforcing Axioms for AI Alignment under Loss-Based RulesPapers · 9 Nov 2026 · Alexandros Hollender, Sonja Kraiczy
- AERO: Adaptive Ensemble-Disagreement Routing for Oracle Feedback in Sample-Efficient Online RLHFPapers · 7 Nov 2026 · Shrey Singhal, Nannan Huang, Haytham M. Fayek
- Goodhart in the Planner: Sampling-Based Control Exploits Rewards Learned from Human PreferencesPapers · 6 Nov 2026 · Soovadeep Bakshi, Apurba Bose
- AUTEUR: Acoustic Plan Learning for Red-Teaming Text-to-Audio ModelsPapers · 6 Nov 2026 · Jiashu Zhang, Jiaming He, Zheming Zhang, Kaizheng Wang, Bing Li, Zhiyuan Zha, Yi…
- Don't Blame The Preferences: Examining The Emergence of Sycophancy in LLM TrainingPapers · 5 Nov 2026 · Chao Peter Yang, Catherine Zehr, Pranav Bijith, Brandon Fain
- Vision Paper: Why Distributed LLM Post-Training and Decentralized Alignment Are Now EssentialPapers · 4 Nov 2026 · Zhanhong Jiang, Qisai Liu, Soumik Sarkar
- Show HN: Interactive notes on RLHF, attention, LoRA and LLM evalsDiscussion · 3 Nov 2026 · laxmena · 41 points
- Correct Robots Before They Make Mistakes: Proactive Human-in-the-Loop Intervention via Preference LearningPapers · 3 Nov 2026 · Xinkai Jiang, Jing Jiang, Hongyi Zhou, Pankhuri Vanjani, Ahmad Baki, Jakub Sulig…
- Tencent-Hunyuan/EmoRewardCode · 2 Nov 2026 · Tencent-Hunyuan · 41 stars
- Forecasting How RLHF Responds to Reward-Model Edits Before TrainingPapers · 2 Nov 2026 · Zixiao Wang, Duzhen Zhang, Xingbo Du, Dongxia Wu, Le Song
- shakewingo/TadpoleCode · 26 Oct 2026 · shakewingo · 96 stars