RLHF

Source: https://github.com/opendilab/awesome-RLHF at 96acdd4571e7
License: Apache-2.0
Read the English original

This edition is an AI-assisted translation.

アウィサム RLHF (人間のフィードバックのRL)

ヒューマンフィードバックで学ぶの補強のための研究論文の集です()RLHF)。 リポジトリは継続的に更新され、フロンティアを追跡します RLHF.

ようこそフォローして星へ!

コンテンツの表

• アウィサム RLHF (人間のフィードバックのRL)
  • コンテンツの表
  • 会社概要 RLHF
    • 詳しい説明
  • ペーパー
    • 2026
    • 2025
    • 2024
    • 2023
    • 2022
    • 2021
    • 2020年以降
  • コードベース
  • データセット
  • ブログ
  • 出版書籍
  • その他の言語サポート
  • 貢献する
  • ライセンス

会社概要 RLHF

コンセプト RLHF 強化学習から直接人間のフィードバックで言語モデルを最適化するためのメソッドを使用することです。 RLHF テキストデータの一般的なコルパスで訓練されたモデルを複雑な人間の値に整列し始める言語モデルを有効にしました。

• RLHF 大きい言語モデルのため()LLM)

• RLHF ビデオゲーム(例) Atari)

詳しい説明

(次のセクションは自動的に生成されます) ChatGPT)

RLHF 一般的に「人間のフィードバックで学習する補強」を参照してください。 強化学習(RL)は、その環境からのフィードバックに基づいて意思決定を行うためにエージェントを訓練することを含む機械学習の一種です。 インスタグラム RLHF, エージェントは、その行動の評価や評価の形で人からのフィードバックを受け取ります, それはより迅速かつ正確に学ぶことができます.

RLHF ロボット、ゲーム、パーソナライズされた推奨システムなどの分野におけるアプリケーションと、人工知能のアクティブな研究領域です。 エージェントが環境からのフィードバックへのアクセスを制限し、その性能を向上させるために人間の入力を必要とするシナリオでRLの課題に対処することを目指しています。

人間のフィードバックで学習する補強 (RLHF)は人工知能の研究の急速に発達した区域であり、性能を改善するために開発された複数の先端技術があります RLHF システム。 ここにいくつかの例があります。

• Inverse Reinforcement Learning (IRL): IRL エージェントは、あらかじめ定義された報酬関数に依存するのではなく、人間のフィードバックから報酬関数を学ぶことができる技術です。 これにより、エージェントは、希望する行動のデモンストレーションなど、より複雑なフィードバック信号から学習することができます。

• Apprenticeship Learning: 実習生が融合する技術 IRL エージェントが人間のフィードバックと専門家のデモの両方から学ぶことを可能にするように指示された学習を使って。 これは、エージェントがプラスとマイナスのフィードバックの両方から学ぶことができるので、より迅速かつ効果的に学習するのに役立ちます。

• Interactive Machine Learning (IML): IML エージェントと人の専門家の間で積極的な相互作用を伴う技術で、エキスパートがエージェントの行動をリアルタイムでフィードバックできるようにします。 エージェントは、学習プロセスの各ステップでそのアクションに対するフィードバックを受信できるように、より迅速かつ効率的に学習するのに役立ちます。

• Human-in-the-Loop Reinforcement Learning (HITLRL): HITLRL リワードシェーピング、アクション選択、ポリシー最適化など、複数のレベルのRLプロセスに人間のフィードバックを統合する技術です。 これは、効率と有効性を改善するのに役立ちます RLHF 人と機械の強みを生かしたシステム。

人間のフィードバックで学習する補強の例は次のとおりです。RLHF):

• Game Playing: ゲームプレイでは、人間のフィードバックは、エージェントが異なるゲームシナリオで有効である戦略と戦術を学ぶのを助けることができます。 たとえば、Goの人気ゲームでは、人間の専門家は、その動きのエージェントにフィードバックを提供でき、ゲームプレイと意思決定を向上させることができます。

• Personalized Recommendation Systems: 推奨システムでは、人的フィードバックは、エージェントが個々のユーザーの好みを学び、パーソナライズされた勧告を提供できるようにするのに役立ちます。 たとえば、エージェントは、ユーザーが推奨する製品からフィードバックを使用して、どの機能がそれらにとって最も重要なかを知ることができます。

• Robotics: ロボティクスでは、人間のフィードバックは、エージェントが安全で効率的な方法で物理的な環境と相互作用する方法を学ぶのを助けることができます。 たとえば、ロボットは新しい環境を素早くナビゲートし、最良の方法やオブジェクトを避けるために、人間のオペレータからのフィードバックでより迅速に移動することができます。

• Education: 教育では、人間のフィードバックは、エージェントがより効果的に学生を教える方法を学ぶのを助けることができます。 たとえば、AIベースのチューターは、さまざまな学生と戦略を教える教師からのフィードバックを使用して、学習経験をパーソナライズするのに役立ちます。

ペーパー

また、 リンクはこちら AI強化紙の読書体験を得る。

format:
- [title](paper link) [links]
  - author1, author2, and author3...
  - publisher
  - keyword
  - code
  - experiment environments and datasets

2026

• Why DPO is a Misspecified Estimator and How to Fix It

  • Aditya Gopalan, Sayak Ray Chowdhury, Debangshu Banerjee
  • キーワード: DPO, RLHF, Preference, Alignment, LLM

• What’s In My Human Feedback? Learning Interpretable Descriptions of Preference Data

  • Rajiv Movva, Smitha Milli, Sewon Min, Emma Pierson
  • キーワード: RLHF, Preference, Alignment, Safety, Human Feedback

• Multiplayer Nash Preference Optimization

  • Fang Wu, Xu Huang, Weihao Xuan, Zhiwei Zhang, Yijia Xiao, Guancheng Wan, Xiaomin Li, Bing Hu, Peng Xia, Jure Leskovec, Yejin Choi
  • キーワード: PPO, RLHF, Preference, Alignment, LLM

• Token-Importance Guided Direct Preference Optimization

  • Ning Yang, Hai Lin, Yibo Liu, Baoliang Tian, Guoqing Liu, Haijun Zhang
  • キーワード: DPO, RLHF, Preference, Alignment, LLM

• SafeDPO: A Simple Approach to Direct Preference Optimization with Enhanced Safety

  • Geon-Hyeong Kim, Yu Jin Kim, Byoungjip Kim, Honglak Lee, Kyunghoon Bae, Youngsoo Jang, Moontae Lee
  • キーワード: DPO, RLHF, Reward Model, Preference, Alignment

• BaseReward: A Strong Baseline for Multimodal Reward Model

  • YiFan Zhang, Haihua Yang, Huanyu Zhang, Yang Shi, Zezhou Chen, Haochen Tian, Chaoyou Fu, Kai WU, Bo Cui, Xu Wang, Jianfei Pan, Haotian Wang, Zhang Zhang, Liang Wang
  • キーワード: RLHF, Reward Model, Preference, Multimodal, LLM

• The Alignment Auditor: A Bayesian Framework for Verifying and Refining LLM Objectives

  • Matthieu Bou, Nyal Patel, Arjun Jagota, Satyapriya Krishna, Sonali Parbhoo
  • キーワード: RLHF, Preference, Alignment, Safety, LLM

• Uni-DPO: A Unified Paradigm for Dynamic Preference Optimization of LLMs

  • Shangpin Peng, Weinong Wang, Zhuotao Tian, Senqiao Yang, Xing W, Haotian Xu, Chengquan Zhang, Takashi Isobe, Baotian Hu, Min Zhang
  • キーワード: DPO, RLHF, Preference, Multimodal, LLM

• Learning to summarize user information for personalized reinforcement learning from human feedback

  • HyunJi Nam, Yanming Wan, Mickel Liu, Peter F. Ahnn, Jianxun Lian, Natasha Jaques
  • キーワード: RLHF, Reward Model, Preference, Alignment, LLM

• Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding

  • Yifan Zhu, Huiqiang Rong, Haoran Luo
  • キーワード: RLHF, LLM, Token-level, Reinforcement Learning, Human Feedback

• Pretrain Value, Not Reward: Decoupled Value Policy Optimization

  • Chenghua Huang, Lu Wang, Fangkai Yang, Pu Zhao, Qingwei Lin, Dongmei Zhang, Saravan Rajmohan
  • キーワード: RLHF, Reward Model, Preference, LLM, Optimization

• P$^2$-DPO: Grounding Hallucination in Perceptual Processing via Calibration Direct Preference Optimization

  • ruipeng zhang, Zhihao Li, Haozhang Yuan, C.L.Philip Chen, Tong Zhang
  • キーワード: DPO, Preference, Optimization, Human Feedback

• Unifying Stable Optimization and Reference Regularization in RLHF

  • Li He, Qiang Qu, He Zhao, Stephen Wan, Dadong Wang, Lina Yao, Tongliang Liu
  • キーワード: RLHF, Preference, Alignment, Optimization, Reinforcement Learning

• Text2Grad: Reinforcement Learning from Natural Language Feedback

  • Hanyang Wang, Lu Wang, Chaoyun Zhang, Tianjun Mao, Si Qin, Qingwei Lin, Saravan Rajmohan, Dongmei Zhang
  • キーワード: RLHF, Reward Model, Alignment

• ARMOR: Aligning Secure and Safe Large Language Models via Meticulous Reasoning

  • Zhengyue Zhao, YingziYingzi Ma, Somesh Jha, Marco Pavone, Patrick McDaniel, Chaowei Xiao
  • キーワード: RLHF, Alignment, Safety, LLM, Optimization

• Reward Model Routing in Alignment

  • Xinle Wu, Yao Lu
  • キーワード: RLHF, Reward Model, Preference, Alignment, LLM

• CogFlow: Bridging Perception and Reasoning through Knowledge Internalization for Visual Mathematical Problem Solving

  • Shuhang Chen, Yunqiu Xu, Junjie Xie, Aojun Lu, Tao Feng, ZEYING HUANG, ZHANG NING, Yi Sun, Yi Yang, Hangjie Yuan
  • キーワード: Reward Model, Multimodal, Optimization

• All Roads Lead to Likelihood: The Value of Reinforcement Learning in Fine-Tuning

  • Gokul Swamy, Sanjiban Choudhury, Wen Sun, Steven Wu, Drew Bagnell
  • キーワード: PPO, Reward Model, Preference, Reinforcement Learning

• General Exploratory Bonus for Optimistic Exploration in RLHF

  • Wendi Li, Changdae Oh, Sharon Li
  • キーワード: RLHF, Alignment, Reinforcement Learning, Human Feedback

• RE-PO: Robust Enhanced Policy Optimization as a General Framework for LLM Alignment

  • Xiaoyang Cao, Zelai Xu, Mo Guang, Kaiwen Long, Michiel A. Bakker, Yu Wang, Chao Yu
  • キーワード: DPO, RLHF, Preference, Alignment, LLM

• Learning Correlated Reward Models: Statistical Barriers and Opportunities

  • Yeshwanth Cherapanamjeri, Constantinos Costis Daskalakis, Gabriele Farina, Sobhan Mohammadpour
  • キーワード: RLHF, Reward Model, Preference, Reinforcement Learning, Human Feedback

• Verification and Co-Alignment via Heterogeneous Consistency for Preference-Aligned LLM Annotations

  • Cheng Chen, Haiyan Yin, Ivor Tsang
  • キーワード: RLHF, Preference, Alignment, LLM

• Disentangling Length Bias in Preference Learning via Response-Conditioned Modeling

  • Jianfeng Cai, Jinhua Zhu, Ruopei Sun, Yue Wang, Li Li, Wengang Zhou, Houqiang Li
  • キーワード: DPO, RLHF, Reward Model, Preference, LLM

• Enforcing Axioms for AI Alignment under Loss-Based Rules

  • Alexandros Hollender, Sonja Kraiczy
  • キーワード: RLHF, Reward Model, Preference, Alignment, Reinforcement Learning

• OPPO: Accelerating PPO-based RLHF via Pipeline Overlap

  • Kaizhuo Yan, YingJie Yu, Yifan Yu, Haizhong Zheng, Fan Lai
  • キーワード: PPO, RLHF, Reward Model, Preference, LLM

• QuRL: Rubrics As Judge For Open-Ended Question Answering

  • Xiyu Wei, Qingwei Zong, Xiaoguang Li, Eugene J. Yu, Sujian Li
  • キーワード: LLM, Optimization, Reinforcement Learning, Human Feedback

• Translate Policy to Language: Flow Matching Generated Rewards for LLM Explanations

  • Xinyi Yang, Liang Zeng, Heng Dong, Chao Yu, Xiaoran Wu, Huazhong Yang, Yu Wang, Milind Tambe, Tonghan Wang
  • キーワード: RLHF, LLM, Reinforcement Learning

• Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy

  • Chris Yuhao Liu, Liang Zeng, Yuzhen Xiao, Jujie He, Jiacai Liu, Chaojie Wang, Rui Yan, Wei Shen, Fuxiang Zhang, Jiacheng Xu, Yang Liu
  • キーワード: RLHF, Reward Model, Preference, Alignment, Safety

• Stackelberg Learning from Human Feedback: Preference Optimization as a Sequential Game

  • Barna Pásztor, Thomas Kleine Buening, Andreas Krause
  • キーワード: RLHF, Preference, Alignment, Nash, Optimization

• Swap-guided Preference Learning for Personalized Reinforcement Learning from Human Feedback

  • Gihoon Kim, Euntai Kim
  • キーワード: RLHF, Reward Model, Preference, Reinforcement Learning, Human Feedback

• Beyond Binary Preferences: A Principled Framework for Reward Modeling with Ordinal Feedback

  • Amirhossein Afsharrad, Ruida Zhou, Luca Viano, Sanjay Lall, Mohammad Ghavamzadeh
  • キーワード: Reward Model, Preference, Safety, Human Feedback

• RLBFF: Binary Flexible Feedback to bridge between Human Feedback & Verifiable Rewards

  • Zhilin Wang, Jiaqi Zeng, Olivier Delalleau, Ellie Evans, Daniel Egert, Hoo-Chang Shin, Felipe Soares, Yi Dong, Oleksii Kuchaiev
  • キーワード: RLHF, Reward Model, Preference, Alignment, LLM

• Reward Models Inherit Value Biases from Pretraining

  • Brian Christian, Jessica A F Thompson, Elle, Vincent Adam, Hannah Rose Kirk, Christopher Summerfield, Tsvetomira Dumbalska
  • キーワード: Reward Model, Preference, Alignment, Safety, LLM

• Semantic-aware Wasserstein Policy Regularization for Large Language Model Alignment

  • Byeonghu Na, Hyungho Na, Yeongmin Kim, Suhyeon Jo, HeeSun Bae, Mina Kang, Il-chul Moon
  • キーワード: RLHF, Preference, Alignment, LLM, Reinforcement Learning

• RewardBench 2: Advancing Reward Model Evaluation

  • Saumya Malik, Valentina Pyatkin, Sander Land, Jacob Morrison, Noah A. Smith, Hannaneh Hajishirzi, Nathan Lambert
  • キーワード: RLHF, Reward Model, Preference, Alignment, Safety

• Causally Robust Reward Learning from Reason-Augmented Preference Feedback

  • Minjune Hwang, Yigit Korkmaz, Daniel Seita, Erdem Biyik
  • キーワード: Reward Model, Preference

• COMAL: A Convergent Meta-Algorithm for Aligning LLMs with General Preferences

  • Yixin Liu, Argyris Oikonomou, Weiqiang Zheng, Yang Cai, Arman Cohan
  • キーワード: RLHF, Preference, Alignment, Nash, Optimization

• Displacement-Resistant Extensions of DPO with Nonconvex $f$-Divergences

  • Idan Pipano, Shoham Sabach, Kavosh Asadi, Mohammad Ghavamzadeh
  • キーワード: DPO, RLHF

• Keep the Best, Forget the Rest: Reliable Alignment with Order-Aware Preference Optimization

  • Jiahui Zhu, Yuanjie Shi, Xiyue Peng, Xin Liu, Yan Yan, Honghao Wei
  • キーワード: DPO, PPO, RLHF, Preference, Alignment

• Cultivating Pluralism In Algorithmic Monoculture: The Community Alignment Dataset

  • Lily H Zhang, Smitha Milli, Karen Long Jusko, Jonathan Smith, Brandon Amos, Wassim Bouaziz, Manon Revel, Jack Kussman, Yasha Sheynin, Lisa Titus, Bhaktipriya Radharapu, Jane Yu, Vidya Sarma, Kristopher Rose, Maximilian Nickel
  • キーワード: Preference, Alignment, LLM

• Fair Reinforcement Learning for Just AI

  • Ezgi Korkmaz
  • キーワード: Preference, Alignment, Optimization, Reinforcement Learning, Human Feedback

• Robust Reward Modeling via Causal Rubrics

  • Pragya Srivastava, Harman Singh, Rahul Madhavan, Gandharv Patil, Sravanti Addepalli, Arun Suggala, Rengarajan Aravamudhan, Soumya Sharma, Anirban Laha, Aravindan Raghuveer, Karthikeyan Shanmugam, Doina Precup
  • キーワード: DPO, Reward Model, Alignment, Safety, LLM

• Escaping Policy Contraction: Contraction-Aware PPO (CaPPO) for Stable Language Model Fine-Tuning

  • Dun Yuan, Di Wu, Xue Liu
  • キーワード: PPO, RLHF, Alignment, Optimization, Reinforcement Learning

• Beyond Pairwise: Empowering LLM Alignment With (Ranked) Choice Modeling

  • Yuxuan Tang, Yifan Feng
  • キーワード: DPO, PPO, Preference, Alignment, LLM

• Learning Ordinal Probabilistic Reward from Preferences

  • Longze Chen, Lu Wang, Renke Shan, Ze Gong, Run Luo, Jiaming Li, Jing Luo, Qiyao Wang, Min Yang
  • キーワード: Reward Model, LLM

• Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance

  • Zhuo Li, Pengyu Cheng, Zhechao Yu, FeifeiTong, Anningzhe Gao, Tsung-Hui Chang, Xiang Wan, erchao.zec, xiaoxi jiang, guanjunjiang
  • キーワード: RLHF, Reward Model, Preference, LLM, Optimization

• Alignment-Weighted DPO: A principled reasoning approach to improve safety alignment

  • Mengxuan Hu, Vivek Datla, Anoop Kumar, Zihan Guan, Sheng Li, Alfy Samuel, Daben Liu
  • キーワード: DPO, RLHF, Preference, Alignment, Safety

• Evaluating and Improving Cultural Awareness of Reward Models for LLM Alignment

  • Hongbin Zhang, Kehai Chen, Xuefeng Bai, Yang Xiang, Min Zhang
  • キーワード: Reward Model, Preference, Alignment, LLM, Reinforcement Learning

• Balancing the Experts: Unlocking LoRA-MoE for GRPO via Mechanism-Aware Rewards

  • Changlian Ma, Zizheng Huang, Xiangyu Zeng, Yi Wang, Cheng Liang, Kun Tian, Xinhai Zhao, Limin Wang
  • キーワード: Alignment, Multimodal, Optimization, Reinforcement Learning

• Bradley-Terry and Multi-Objective Reward Modeling Are Complementary

  • Zhiwei Zhang, Hui Liu, Xiaomin Li, Zhenwei Dai, Jingying Zeng, Fali Wang, Minhua Lin, Ramraj Chandradevan, Linlin Wu, Zhen Li, Chen Luo, Zongyu Wu, Xianfeng Tang, Qi He, Suhang Wang
  • キーワード: RLHF, Reward Model, Preference, LLM, Reinforcement Learning

• Beyond RLHF and NLHF: Population-Proportional Alignment under an Axiomatic Framework

  • Kihyun Kim, Jiawei Zhang, Asuman E. Ozdaglar, Pablo A. Parrilo
  • キーワード: Preference, Alignment

• ActiveDPO: Active Direct Preference Optimization for Sample-Efficient Alignment

  • Xiaoqiang Lin, Arun Verma, Zhongxiang Dai, Daniela Rus, See-Kiong Ng, Bryan Kian Hsiang Low
  • キーワード: DPO, Reward Model, Preference, Alignment, LLM

• BranchGRPO: Stable and Efficient GRPO with Structured Branching in Diffusion Models

  • Yuming Li, Yikai Wang, Yuying zhu, Zhongyu Zhao, Ming Lu, Qi She, Shanghang Zhang
  • キーワード: Preference, Alignment, Optimization

• Safety Game: Inference-Time Alignment of Black-Box LLMs via Constrained Optimization

  • Tuan Nguyen, Long Tran-Thanh
  • キーワード: Alignment, Safety, LLM, Reinforcement Learning, Human Feedback

• Threshold-Guided Optimization for Visual Generative Models

  • Jinbin Bai, Yu Lei, Qingyu Shi, Aosong Feng, Yi Xin, Zhuoran Zhao, Fei Shen, Kaidong Yu, Xiangtai Li
  • キーワード: Reward Model, Preference, Alignment, Diffusion, Optimization

• Noise-corrected GRPO: From Noisy Rewards to Unbiased Gradients

  • Omar Elmansouri, Fathinah Izzati, Mohamed El Amine Seddik, Salem Lahlou
  • キーワード: RLHF, Reward Model, LLM, Optimization, Reinforcement Learning

• Controllable and explainable personality sliders for LLMs at inference time

  • Florian Hoppe, David Khachaturov, Robert Mullins, Mark Huasong Meng
  • キーワード: PPO, RLHF, Alignment, LLM, Optimization

• PS-PPO : Prefix-Sampling PPO for Critic-Free RLHF

  • Doo Hwan Hwang, Kee-Eung Kim
  • キーワード: PPO, RLHF, Optimization, Reinforcement Learning, Human Feedback

• Unbiased Reward Modeling from Implicit Preference

  • Eric Wang, Haocheng Yang, Licheng Pan, Lei Shen, Xiaoxi Li, Yinuo Wang, Zhichao Chen, Yuan Lu, Haoxuan Li, Zhouchen Lin
  • キーワード: RLHF, Reward Model, Preference, Reinforcement Learning, Human Feedback

• How RLHF Amplifies Sycophancy

  • Itai Shapira, Gerdus Benade, Ariel Procaccia
  • キーワード: Preference, Alignment, Optimization, Human Feedback

• Real-Time Aligned Reward Model beyond Semantics

  • Zixuan Huang, Xin Xia, Yuxi Ren, Jianbin Zheng, Xuefeng Xiao, Hongyan Xie, Huaqiu Li, Songshi Liang, Zhongxiang Dai, Fuzhen Zhuang, Jianxin Li, Yikun Ban, deqing wang
  • キーワード: RLHF, Reward Model, Preference, Alignment, LLM

• Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Models

  • Mickel Liu, Liwei Jiang, Yancheng Liang, Simon Du, Yejin Choi, Tim Althoff, Natasha Jaques
  • キーワード: RLHF, Reward Model, Alignment, Safety, LLM

• B-Spar: Bayesian Sparse-Reward Modeling for RL-based Image Editing

  • shusong xu, Peiye Liu, Yongbin Liu, Bangjie Yin, Zhaomang Sun, Zhenyu Chen, Tianyi Zheng, Peng-Tao Jiang, Jian Zhang, Yuzhao Wang, Jinwei Chen, Zhen Gu, Bo Li
  • キーワード: Reward Model, Alignment, Multimodal, LLM, Optimization

• Calibrated Preference Learning: The Case of Label Ranking

  • Santo Thies, Viktor Bengs, Timo Kaufmann, Sebastian Vollmer, Eyke Hüllermeier
  • キーワード: RLHF, Reward Model, Alignment

• Understanding the Performance Gap in Preference Learning: A Dichotomy of RLHF and DPO

  • Ruizhe Shi, Minhak Song, Runlong Zhou, Zihan Zhang, Maryam Fazel, Simon Du
  • キーワード: DPO, RLHF, Reward Model, Preference, Optimization

• DARC: Disagreement-Aware Alignment via Risk-Constrained Decoding

  • mingxi Zou, Jiaxiang Chen, Junfan Li, Langzhang Liang, Qifan Wang, Xu Yinghui, Zenglin Xu
  • キーワード: DPO, RLHF, Preference, Alignment, Optimization

• The Personality Illusion: Revealing Dissociation Between Self-Reports & Behavior in LLMs

  • Pengrui Han, Rafal Kocielnik, Peiyang Song, Ramit Debnath, Dean Mobbs, Anima Anandkumar, R. Michael Alvarez
  • キーワード: RLHF, Alignment, LLM

• Distributionally Robust Reinforcement Learning with Human Feedback

  • Debmalya Mandal, Paulius Sasnauskas, Goran Radanovic
  • キーワード: DPO, RLHF, Reward Model, Preference, LLM

• Automatically Finding Reward Model Biases

  • Atticus Wang, Iván Arcuschin, Arthur Conmy
  • キーワード: Reward Model, LLM, Reinforcement Learning, Human Feedback

• Tackling Length Inflation Without Trade-offs: Group Relative Reward Rescaling for Reinforcement Learning

  • Zichao Li, Jie Lou, Fangchen Dong, Zhiyuan Fan, Mengjie Ren, Hongyu Lin, Xianpei Han, Debing Zhang, Le Sun, Yaojie Lu, XingYu
  • キーワード: RLHF, LLM, Optimization, Reinforcement Learning

• Convex Optimization for Alignment and Preference Learning on a Single GPU

  • Miria Feng, Mert Pilanci
  • キーワード: DPO, RLHF, Preference, Alignment, LLM

• MMKU-Bench: A Multimodal Update Benchmark for Diverse Visual Knowledge

  • Baochen Fu, Yuntao Du, Cheng Chang, Baihao Jin, Wenzhi Deng, Muhao Xu, Hongmei Yan, Weiye Song, Yi Wan
  • キーワード: RLHF, Multimodal, Reinforcement Learning, Human Feedback

• Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization

  • Yihang Yao, Zhepeng Cen, Haohong Lin, Shiqi Liu, Zuxin Liu, Jiacheng Zhu, Zhang-Wei Hong, Laixi Shi, Ding Zhao
  • キーワード: LLM, Reinforcement Learning, Human Feedback

• Unbiased Alignment for Large Language Models with Noisy Preferences

  • Jialiang Wang, Xianming Liu, Xiong Zhou, Hui Liu, Haoliang Li
  • キーワード: DPO, Reward Model, Preference, Alignment, Optimization

• Unbiased Principles, Robust Rewards

  • Qingnan Ren, Zhen Fang, Shiting Huang, Yu Zeng, Lin Chen, Zehui Chen, Feng Zhao
  • キーワード: RLHF, Reward Model, Reinforcement Learning, Human Feedback

• The Secret Engine Behind RLHF: It’s Contarstive Learning All Along

  • Xufei Lv, Kehai Chen, Haoyuan Sun, Xuefeng Bai, Min zhang, Houde Liu
  • キーワード: DPO, RLHF, Preference, Alignment, LLM

• When Distance Distracts: Representation Distance Bias in BT-Loss for Reward Models

  • Tong Xie, Ching-Yuan Bai, Yuanhao Ban, Yunqi Hong, Haoyu Li, Cho-Jui Hsieh
  • キーワード: RLHF, Reward Model, Alignment, LLM

• Multi-Objective Preference Optimization: Improving Human Alignment of Generative Models

  • Akhil Agnihotri, Rahul Jain, Deepak Ramachandran, Zheng Wen
  • キーワード: DPO, RLHF, Preference, Alignment, Safety

• TUR-DPO: Topology- and Uncertainty-Aware Direct Preference Optimization

  • Abdulhady abas, Fatemeh Daneshfar, Seyedali Mirjalili, Mourad Oussalah
  • キーワード: DPO, PPO, RLHF, Preference, Multimodal

• Asymptotic Universal Alignment: A New Alignment Framework via Test-Time Scaling

  • Yang Cai, Weiqiang Zheng
  • キーワード: PPO, Preference, Alignment, LLM, Nash

• Reward Modeling from Natural Language Human Feedback

  • Zongqi Wang, Rui Wang, Yuchuan Wu, Yiyao Yu, Pinyi Zhang, Shaoning Sun, Yujiu Yang, Yongbin Li
  • キーワード: Reward Model, Preference, Reinforcement Learning, Human Feedback

• Efficient Preference Poisoning Attack on Offline RLHF

  • Chenye Yang, Weiyu Xu, Lifeng Lai
  • キーワード: DPO, RLHF, Preference, Optimization, Reinforcement Learning

• Position: Agentic Safety is an Epistemic Property, Not a Behavioral One

  • Charles Wang, Keir Dorchen, Peter Jin
  • キーワード: RLHF, Preference, Alignment, Safety, Optimization

• Position: Large Language Models Should Learn Personalized Rather Than Aggregated Human Preferences

  • Cristina Garbacea
  • キーワード: RLHF, Reward Model, Preference, Safety, Reinforcement Learning

• Transitivity Meets Cyclicity: Explicit Preference Decomposition for Dynamic Large Language Model Alignment

  • Yucong Huang, Xiucheng Li, Kaiqi Zhao, Jing Li
  • キーワード: PPO, RLHF, Preference, Alignment, Nash

• Factored Causal Representation Learning for Robust Reward Modeling in RLHF

  • Yupei Yang, Lin Yang, Wanxi Deng, Lin Qu, Fan Feng, Biwei Huang, Shikui Tu, Lei Xu
  • キーワード: RLHF, Reward Model, Preference, LLM, Reinforcement Learning

• Federated Variational Preference Alignment with Gumbel-Softmax Prior for Personalized User Preferences

  • Jabin Koo, Hoyoung Kim, Minwoo Jang, Jungseul Ok
  • キーワード: RLHF, Reward Model, Preference, Alignment, LLM

• Online Compatible Reward Identification from Preference Feedback

  • Simone Drago, Marco Mussi, Alberto Maria Metelli
  • キーワード: Preference, Safety, Reinforcement Learning, Human Feedback

• $f$-Divergence Regularized RLHF: Two Tales of Sampling and Unified Analyses

  • Di Wu, Chengshuai Shi, Jing Yang, Cong Shen
  • キーワード: RLHF, Reinforcement Learning, Human Feedback

• Alignment Tampering: How Reinforcement Learning from Human Feedback Is Exploited to Optimize Misaligned Biases

  • Dongyoon Hahm, Dylan Hadfield-Menell, Kimin Lee
  • キーワード: RLHF, Reward Model, Preference, Alignment, LLM

• IRPM: Intergroup Relative Preference Modeling for Pointwise Generative Reward Models

  • Haonan Song, Qingchen Xie, Huan Zhu, Feng Xiao, Luxi Xing, Liu Kang, Fuzhen Li, Zhiyong Zheng, Feng Jiang, Ziheng Li, Kun Yan, Qingyi Si, Yanghua Xiao, Hongcheng Guo, Fan Yang
  • キーワード: RLHF, Reward Model, Preference, Reinforcement Learning, Human Feedback

• Implicit Preference Alignment for Human Image Animation

  • Yuanzhi Wang, Xuhua Ren, Jiaxiang Cheng, bing ma, Kai Yu, Tianxiang Zheng, Qinglin Lu, Zhen Cui
  • キーワード: Preference, Alignment, Optimization, Reinforcement Learning, Human Feedback

• Multilingual Safety Alignment Via Sparse Weight Editing

  • Jiaming Liang, Zhaoxin Wang, Handing Wang
  • キーワード: RLHF, Alignment, Safety, LLM, Reinforcement Learning

• Gradient Regularization Prevents Reward Hacking in Reinforcement Learning from Human Feedback and Verifiable Rewards

  • Johannes Ackermann, Michael Noukhovitch, Takashi Ishida, Masashi Sugiyama
  • キーワード: RLHF, Reward Model, LLM, Reinforcement Learning, Human Feedback

• Graph-Preference Learning: Debiasing Network-Sampled Human Feedback for Target Welfare Estimation

  • Guangrui Fan, DanDan Liu, AZNUL SABRI, Pan Lihu
  • キーワード: DPO, RLHF, Reward Model, Preference

• COLLIE: Guiding Skill Discovery in Semantically Coherent Latent Space

  • Yao Luan, Ni Mu, Hanfei Ge, Yiqin Yang, Bo XU, Qing-Shan Jia
  • キーワード: Human Feedback

• Optimal Transport for Reward Modeling from Noisy Feedback

  • Eric Wang, Licheng Pan, Haocheng Yang, Yunsheng Lu, Yongqi Tong, Yinuo Wang, Shijian Wang, Zhixuan Chu, Lei Shen, Haoxuan Li, Yuan Lu
  • キーワード: RLHF, Reward Model, Preference, Reinforcement Learning, Human Feedback

• Reliability-Aware LLM Alignment from Inconsistent Human Feedback

  • Jingyi Huang, Ruohan Zong, Yujun Feng, Liran Ma, Lanyu Shang, Yang Zhang
  • キーワード: DPO, RLHF, Preference, Alignment, LLM

• Position: We Need Large Language Models Optimized For Our Well-Being

  • Ashton Anderson, Harsh Kumar, Louis Tay, Karina Vold
  • キーワード: RLHF, Preference, Optimization, Reinforcement Learning, Human Feedback

• Implicit Safety Alignment from Crowd Preferences

  • Qian Lin, Daniel S Brown
  • キーワード: RLHF, Reward Model, Preference, Safety, Reinforcement Learning

• Contrastive Weak-to-Strong Generalization

  • Houcheng Jiang, Junfeng Fang, Jiaxin Wu, Tianyu Zhang, Chen Gao, Xiang Wang, Xiangnan He, Yang Deng
  • キーワード: Reward Model, Alignment, LLM, Human Feedback

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

  • Kazusato Oko, Annie Ulichney, Nika Haghtalab, Han Bao
  • キーワード: RLHF, Preference, Reinforcement Learning, Human Feedback

• Conversation for Non-verifiable Learning: Self-Evolving Large Language Models through Meta-Evaluation

  • Yuan Sui, Bryan Hooi
  • キーワード: LLM, Optimization, Human Feedback

• Unifying Adversarial Robustness and Training Across Text Scoring Models

  • Manveer Tamber, Hosna Oyarhoseini, Jimmy Lin
  • キーワード: PPO, RLHF, Reward Model, LLM

• ActiveUltraFeedback: Efficient Preference Data Generation using Active Learning

  • Davit Melikidze, Marian Schneider, Jessica Lam, Martin Wertich, Ido Hakimi, Barna Pasztor, Andreas Krause
  • キーワード: RLHF, Preference, Alignment, LLM, Reinforcement Learning

• Layer-wise Gradient Disentanglement: Decoupling Semantics and Preferences in Direct Preference Optimization

  • Mengyang Li, Shuang Liu, Zhong Zhang
  • キーワード: DPO, RLHF, Preference, Optimization

• The Sign Estimator: Preference Modeling for LLM Alignment under Heterogeneity

  • Aymane El Gadarri, Ali Aouad, Vivek Farias
  • キーワード: RLHF, Reward Model, Preference, Alignment, LLM

• Leveraging Machine Unlearning for Cost-Efficient Preference Alignment

  • XiaoHua Feng, Yuyuan Li, HuWei Ji, Li Zhang, Jiaming Zhang, Tianyu Du, Chaochao Chen
  • キーワード: Preference, Alignment, LLM, Optimization, Reinforcement Learning

• Regularization in the Axiomatic Approach to Learning from Human Preferences

  • Ezgi Korkmaz
  • キーワード: RLHF, Preference, Reinforcement Learning, Human Feedback

• DPO Unchained: Your Training Algorithm is Secretly Disentangled in Human Choice Theory (and Its Loss’ Convexity is Dispensable)

  • Wenxuan Zhou, Shujian Zhang, brice magdalou, John Lambert, Ehsan Amid, Richard Nock, Andrew Hard
  • キーワード: DPO, PPO, RLHF, Reward Model, Preference

• Conditional Equivalence of DPO and RLHF: Assumptions, Failure Modes, and Provable Alignment

  • Yonggang Zhang, Zhiqin Yang, Wei Xue, Dong Fang, Bo Han, Yike Guo
  • キーワード: DPO, RLHF, Preference, Alignment, Optimization

• A Regret Minimization Framework on Preference Learning in Large Language Models

  • Suhwan Kim, Taehyun Cho, Youngsoo Jang, Geon-Hyeong Kim, Yu Jin Kim, Moontae Lee, Jungwoo Lee
  • キーワード: RLHF, Preference, Optimization, Reinforcement Learning, Human Feedback

• Position: Measuring Human Preferences in RLHF is a Social Science Problem

  • Bijean Ghafouri, Eun Cheol Choi, Priyanka Dey, Emilio Ferrara
  • キーワード: RLHF, Preference, Alignment

• Mitigating Reward Hacking in RLHF via Bayesian Non-negative Reward Modeling

  • Zhibin Duan, Guowei Rong, Zhuo Li, Bo Chen, Mingyuan Zhou, Dandan Guo
  • キーワード: Reward Model, Preference, LLM, Optimization, Reinforcement Learning

2025

• Position: The Complexity of Perfect AI Alignment — Formalizing the RLHF Trilemma

  • Subramanyam Sahoo, Aman Chadha, Vinija Jain, Divya Chaudhary
  • キーワード: Alignment Bias, Safety, Interpretability

• What’s In My Human Feedback? Learning Interpretable Descriptions of Preference Data

  • Rajiv Movva, Smitha Milli, Sewon Min, Emma Pierson
  • キーワード: Sparse Autoencoders, Interpretable Data Curation, Reward Hacking, Feature Attribution
  • コード: Official

• Towards Efficient Online Exploration for Reinforcement Learning with Human Feedback

  • Gen Li, Yuling Yan
  • キーワード: Online RL, Multi-armed Bandit, LLMs

• OpenRLHF: A Ray-based Easy-to-use, Scalable and High-performance RLHF Framework

  • Jian Hu, Xibin Wu, Wei Shen, Jason Klein Liu, Weixun Wang, Songlin Jiang, Haoran Wang, Hao Chen, Bin Chen, Wenkai Fang, Xianyu, Yu Cao, Haotian Xu, Yiming Liu
  • キーワード: Framework
  • コード: Official

• Language Models Learn to Mislead Humans via RLHF

  • Jiaxin Wen, Ruiqi Zhong, Akbir Khan, Ethan Perez, Jacob Steinhardt, Minlie Huang, Sam Bowman, He He, Shi Feng
  • キーワード: Open-ended Task, Human Reward, Alignment Method, LLMs

• A Simple and Effective Reinforcement Learning Method for Text-to-Image Diffusion Fine-tuning

  • Shashank Gupta, Chaitanya Ahuja, Tsung-Yu Lin, Sreya Dutta Roy, Harrie Oosterhuis, Maarten de Rijke, and Satya Narayan Shukla
  • キーワード: Diffusion Model, REINFORCE, PPO

• Differential Information: An Information-Theoretic Perspective on Preference Optimization

  • Yunjae Won, Hyunji Lee, Hyeonbin Hwang, Minjoon Seo
  • キーワード: Preference Optimization, Information-Theoretic Analysis, Log-Ratio Reward Parameterization, Data Distribution, Log-Likelihood Displacement

• Generalist Reward Models: Found Inside Large Language Models

  • Yi-Chen Li, Tian Xu, Yang Yu, Xuqin Zhang, Xiong-Hui Chen, Zhongxiang Ling, Ningjing Chao, Lei Yuan, Zhi-Hua Zhou
  • キーワード: Offline Inverse RL, LLM-as-a-judge, Training-free, Alignment

• A Unified Pairwise Framework for RLHF: Bridging Generative Reward Modeling and Policy Optimization

  • Wenyuan Xu, Xiaochen Zuo, Chao Xin, Yu Yue, Lin Yan, Yonghui Wu
  • キーワード: Generative Pairwise Reward Model, Policy Optimization, Framework

• Exploring Data Scaling Trends and Effects in Reinforcement Learning from Human Feedback

  • Wei Shen, Guanlin Liu, Zheng Wu, Ruofei Zhu, Qingping Yang, Chao Xin, Yu Yue, Lin Yan
  • キーワード: Data Scaling, Reward Hacking, LLMs

• RLTHF: Targeted Human Feedback for LLM Alignment

  • Yifei Xu, Tusher Chakraborty, Emre Kıcıman, Bibek Aryal, Eduardo Rodrigues, Srinagesh Sharma, Roberto Estevao, Maria Angels de Luis Balaguer, Jessica Wolk, Rafael Padilha, Leonardo Nunes, Shobana Balakrishnan, Songwu Lu, Ranveer Chandra
  • キーワード: Human-AI Hybrid Framework, Efficient, Alignment, LLMs

• Equilibrate RLHF: Towards Balancing Helpfulness-Safety Trade-off in Large Language Models

  • Yingshui Tan, Yilei Jiang, Yanshi Li, Jiaheng Liu, Xingyuan Bu, Wenbo Su, Xiangyu Yue, Xiaoyong Zhu, Bo Zheng
  • キーワード: Safety, Framework, Adaptive Message-wise Alignment Method, LLMs

• MM-RLHF: The Next Step Forward in Multimodal LLM Alignment

  • Yi-Fan Zhang, Tao Yu, Haochen Tian, Chaoyou Fu, Peiyan Li, Jianshu Zeng, Wulin Xie, Yang Shi, Huanyu Zhang, Junkang Wu, Xue Wang, Yibo Hu, Bin Wen, Fan Yang, Zhang Zhang, Tingting Gao, Di Zhang, Liang Wang, Rong Jin, Tieniu Tan
  • キーワード: Critique-based Reward Model, Dynamic Reward, Dataset
  • コード: Official

• Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback

  • Yafu Li, Xuyang Hu, Xiaoye Qu, Linjie Li, Yu Cheng
  • キーワード: Test-Time Optimization, Preference Learning, Iterative Feedback
  • コード: Official

• Segmenting Text and Learning Their Rewards for Improved RLHF in Language Model

  • Yueqin Yin, Shentao Yang, Yujia Xie, Ziyi Yang, Yuting Sun, Hany Awadalla, Weizhu Chen, and Mingyuan Zhou
  • キーワード: Segment-level Reward Model, Dense Reward RLHF Framework, Improved PPO training for LLMs
  • コード: Official

• REINFORCE++: A Simple and Efficient Approach for Aligning Large Language Models

  • Jian Hu
  • キーワード: Efficient, Alignment, Reinforcement Learning
  • コード: Official

2024

• DPO Meets PPO: Reinforced Token Optimization for RLHF

  • Han Zhong, Zikang Shan, Guhao Feng, Wei Xiong, Xinle Cheng, Li Zhao, Di He, Jiang Bian, Liwei Wang
  • キーワード: Token-wise Reward, DPO, PPO, RLHF
  • コード: Official

• Reward-Augmented Data Enhances Direct Preference Alignment of LLMs

  • Shenao Zhang, Zhihan Liu, Boyi Liu, Yufeng Zhang, Yingxiang Yang, Yongfei Liu, Liyu Chen, Tao Sun, Zhaoran Wang
  • キーワード: Reward-Augmented Data, DPO, LLMs
  • コード: Official

• The Accuracy Paradox in RLHF: When Better Reward Models Don’t Yield Better Language Models

  • Yanjun Chen, Dawei Zhu, Yirong Sun, Xinghao Chen, Wei Zhang, Xiaoyu Shen
  • キーワード: Reward Model Evaluation, Accuracy Paradox, LLM Alignment
  • コード: Official

• Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback

  • Jiaming Ji, Jiayi Zhou, Hantao Lou, Boyuan Chen, Donghai Hong, Xuyao Wang, Wenqi Chen, Kaile Wang, Rui Pan, Jiahao Li, Mohan Wang, Josef Dai, Tianyi Qiu, Hua Xu, Dong Li, Weipeng Chen, Jun Song, Bo Zheng, Yaodong Yang
  • キーワード: Multi-modality Alignment, Dataset, Training-evaluation Framework
  • コード: Official

• REvolve: Reward Evolution with Large Language Models using Human Feedback

  • Rishi Hazra, Alkis Sygkounas, Andreas Persson, Amy Loutfi, Pedro Zuidberg Dos Martires
  • キーワード: Improved Reward Model with LLMs, Framework
  • コード: Official

• Zeroth-Order Policy Gradient for Reinforcement Learning from Human Feedback without Reward Inference

  • Qining Zhang, Lei Ying
  • キーワード: Reward inference-free RLHF, Zeroth-order optimization, Policy gradient

• Learning Reward and Policy Jointly from Demonstration and Preference Improves Alignment

  • Chenliang Li, Siliang Zeng, Zeyi Liao, Jiaxiang Li, Dongyeop Kang, Alfredo Garcia, Mingyi Hong
  • キーワード: Joint Reward and Policy, Efficiency, Framework

• MA-RLHF: Reinforcement Learning from Human Feedback with Macro Actions

  • Yekun Chai, Haoran Sun, Huang Fang, Shuohuan Wang, Yu Sun, Hua Wu
  • キーワード: Macro action-level Reward, Efficiency, Framework
  • コード: Official

• Reward Modeling with Ordinal Feedback: Wisdom of the Crowd

  • Shang Liu, Yu Pan, Guanting Chen, and Xiaocheng Li
  • キーワード: Reward Modeling, Ordinal Feedback, Human Preference Dataset
  • コード: Official

• Aligning Few-Step Diffusion Models with Dense Reward Difference Learning

  • Ziyi Zhang, Li Shen, Sen Zhang, Deheng Ye, Yong Luo, Miaojing Shi, Bo Du, Dacheng Tao
  • キーワード: Diffusion Models, Text-to-Image, Alignment, Reinforcement Learning
  • コード: Official

• HybridFlow: A Flexible and Efficient RLHF Framework

  • Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, Chuan Wu
  • キーワード: Flexible, Efficient, RLHF framework
  • コード: Official

• ALaRM: Align Language Models via Hierarchical Rewards Modeling

  • Yuhang Lai, Siyuan Wang, Shujun Liu, Xuanjing Huang, Zhongyu Wei
  • キーワード: Hierarchical Reward, Open Text Generation Tasks
  • コード: Official

• TLCR: Token-Level Continuous Reward for Fine-grained Reinforcement Learning from Human Feedback

  • Eunseop Yoon, Hee Suk Yoon, SooHwan Eom, Gunsoo Han, Daniel Wontae Nam, Daejin Jo, Kyoung-Woon On, Mark A. Hasegawa-Johnson, Sungwoong Kim, Chang D. Yoo
  • キーワード: Token-Level Continuous Reward, RLHF
  • コード: Official

• Aligning Large Multimodal Models with Factually Augmented RLHF

  • Zhiqing Sun, Sheng Shen, Shengcao Cao, Haotian Liu, Chunyuan Li, Yikang Shen, Chuang Gan, Liang-Yan Gui, Yu-Xiong Wang, Yiming Yang, Kurt Keutzer, Trevor Darrell
  • キーワード: Factually Augmented RLHF, Vision & Language, Human Preference Dataset
  • コード: Official

• Direct Large Language Model Alignment Through Self-Rewarding Contrastive Prompt Distillation

  • Aiwei Liu, Haoping Bai, Zhiyun Lu, Xiang Kong, Simon Wang, Jiulong Shan, Meng Cao, Lijie Wen
  • キーワード: Without Human Preference Data, Self-Reward, DPO
  • コード: Official

• Arithmetic Control of LLMs for Diverse User Preferences: Directional Preference Alignment with Multi-Objective Rewards

  • Haoxiang Wang, Yong Lin, Wei Xiong, Rui Yang, Shizhe Diao, Shuang Qiu, Han Zhao, Tong Zhang
  • キーワード: User Preference, Multi-objective Reward Model, Rejection Sampling Finetuning
  • コード: Official

• Back to Basics: Revisiting REINFORCE Style Optimization for Learning from Human Feedback in LLMs

  • Arash Ahmadian, Chris Cremer, Matthias Gallé, Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet Üstün, Sara Hooker
  • キーワード: Online RL Optimization, Low Computational Cost
  • コード: Official

• Improving Large Language Models via Fine-grained Reinforcement Learning with Minimum Editing Constraint

  • Zhipeng Chen, Kun Zhou, Wayne Xin Zhao, Junchen Wan, Fuzheng Zhang, Di Zhang, Ji-Rong Wen
  • キーワード: Token-level Reward, LLM
  • コード: Official

• RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback

  • Harrison Lee, Samrat Phatale, Hassan Mansoor, Thomas Mesnard, Johan Ferret, Kellie Ren Lu, Colton Bishop, Ethan Hall, Victor Carbune, Abhinav Rastogi, Sushant Prakash
  • キーワード: RL from AI Feedback
  • コード: official

• Principled Penalty-based Methods for Bilevel Reinforcement Learning and RLHF

  • Han Shen, Zhuoran Yang, Tianyi Chen
  • キーワード: Bilevel optimization
  • コード: official

• Dense Reward for Free in Reinforcement Learning from Human Feedback

  • Alex James Chan, Hao Sun, Samuel Holt, Mihaela Van Der Schaar
  • キーワード: reward shaping, RLHF
  • コード: official

• A Minimaximalist Approach to Reinforcement Learning from Human Feedback

  • Gokul Swamy, Christoph Dann, Rahul Kidambi, Steven Wu, Alekh Agarwal
  • キーワード: Minimax Winner, Self-Play Preference Optimization
  • コード: official

• Rlhf-v: Towards trustworthy mllms via behavior alignment from fine-grained correctional human feedback

  • Tianyu Yu, Yuan Yao, Haoye Zhang, Taiwen He, Yifeng Han, Ganqu Cui, Jinyi Hu, Zhiyuan Liu, Hai-Tao Zheng, Maosong Sun, Tat-Seng Chua
  • キーワード: Multimodal Large Language Models, Hallucination Problem, Reinforcement Learning from Human Feedback
  • コード: official

• RLHF Workflow: From Reward Modeling to Online RLHF

  • Hanze Dong, Wei Xiong, Bo Pang, Haoxiang Wang, Han Zhao, Yingbo Zhou, Nan Jiang, Doyen Sahoo, Caiming Xiong, Tong Zhang
  • キーワード: Online Iterative RLHF, Preference Modeling, Large Language Models
  • コード: official

• MaxMin-RLHF: Towards equitable alignment of large language models with diverse human preferences

  • Souradip Chakraborty, Jiahao Qiu, Hui Yuan, Alec Koppel, Furong Huang, Dinesh Manocha, Amrit Singh Bedi, Mengdi Wang
  • キーワード: mixture of preference distributions, MaxMin alignment objective
  • コード: official

• Dataset Reset Policy Optimization for RLHF

  • Jonathan D. Chang, Wenhao Zhan, Owen Oertell, Kianté Brantley, Dipendra Misra, Jason D. Lee, Wen Sun
  • キーワード: Dataset Reset Policy Optimization
  • コード: official

• A Dense Reward View on Aligning Text-to-Image Diffusion with Preference

  • Shentao Yang, Tianqi Chen, Mingyuan Zhou
  • キーワード: RLHF for Text-to-Image Generation, Dense Reward Improvement of DPO, Efficient Alignment
  • コード: official

• Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models

  • Zixiang Chen, Yihe Deng, Huizhuo Yuan, Kaixuan Ji, Quanquan Gu
  • キーワード: Self-Play Fine-Tuning
  • コード: official

• RLHF Deciphered: A Critical Analysis of Reinforcement Learning from Human Feedback for LLMs

  • Shreyas Chaudhari, Pranjal Aggarwal, Vishvak Murahari, Tanmay Rajpurohit, Ashwin Kalyan, Karthik Narasimhan, Ameet Deshpande, Bruno Castro da Silva
  • キーワード: RLHF, Oracular Reward, Reward Model Analysis, Survey

• Confronting Reward Overoptimization for Diffusion Models: A Perspective of Inductive and Primacy Biases

  • Ziyi Zhang, Sen Zhang, Yibing Zhan, Yong Luo, Yonggang Wen, Dacheng Tao
  • キーワード: Diffusion Models, Alignment, Reinforcement Learning, RLHF, Reward Overoptimization, Primacy Bias
  • コード: official

• On Diversified Preferences of Large Language Model Alignment

  • Dun Zeng, Yong Dai, Pengyu Cheng, Tianhao Hu, Wanshun Chen, Nan Du, Zenglin Xu
  • キーワード: Aligning shared preference, Reward modeling metrics, LLM
  • コード: official

• Aligning Crowd Feedback via Distributional Preference Reward Modeling

  • Dexun Li, Cong Zhang, Kuicai Dong, Derrick Goh Xin Deik, Ruiming Tang, Yong Liu
  • キーワード: RLHF, Preference distribution, Aligning, LLM

• Beyond One-Preference-Fits-All Alignment: Multi-Objective Direct Preference Optimization

  • Zhanhui Zhou, Jie Liu, Chao Yang, Jing Shao, Yu Liu, Xiangyu Yue, Wanli Ouyang, Yu Qiao
  • キーワード: Multi-objective RLHF without reward modeling, DPO
  • コード: official

• Emulated Disalignment: Safety Alignment for Large Language Models May Backfire!

  • Zhanhui Zhou, Jie Liu, Zhichen Dong, Jiaheng Liu, Chao Yang, Wanli Ouyang, Yu Qiao
  • キーワード: LLM inference-time attack, DPO, Producing harmful LLMs without training
  • コード: official

• A Theoretical Analysis of Nash Learning from Human Feedback under General KL-Regularized Preference

  • Chenlu Ye, Wei Xiong, Yuheng Zhang, Nan Jiang, Tong Zhang
  • キーワード: Game-based RLHF, Nash Learning, Alignment under reward-model-free oracle

• Mitigating the Alignment Tax of RLHF

  • Yong Lin, Hangyu Lin, Wei Xiong, Shizhe Diao, Jianmeng Liu, Jipeng Zhang, Rui Pan, Haoxiang Wang, Wenbin Hu, Hanning Zhang, Hanze Dong, Renjie Pi, Han Zhao, Nan Jiang, Heng Ji, Yuan Yao, Tong Zhang
  • キーワード: RLHF, Alignment tax, Catastrophic forgetting

• Training Diffusion Models with Reinforcement Learning

  • Kevin Black, Michael Janner, Yilun Du, Ilya Kostrikov, Sergey Levine
  • キーワード: reinforcement learning, RLHF, diffusion models
  • コード: official

• AlignDiff: Aligning Diverse Human Preferences via Behavior-Customisable Diffusion Model

  • Zibin Dong, Yifu Yuan, Jianye Hao, Fei Ni, Yao Mu, Yan Zheng,Yujing Hu, Tangjie Lv, Changjie Fan, Zhipeng Hu
  • キーワード: Reinforcement learning; Diffusion models; RLHF; Preference aligning
  • コード: official

• Dense Reward for Free in Reinforcement Learning from Human Feedback

  • Alex J. Chan, Hao Sun, Samuel Holt, Mihaela van der Schaar
  • キーワード: RLHF
  • コード: official

• Transforming and Combining Rewards for Aligning Large Language Models

  • Zihao Wang, Chirag Nagpal, Jonathan Berant, Jacob Eisenstein, Alex D’Amour, Sanmi Koyejo, Victor Veitch
  • キーワード: RLHF, Aligning, LLM

• Parameter Efficient Reinforcement Learning from Human Feedback

  • Hakim Sidahmed, Samrat Phatale, Alex Hutcheson, Zhuonan Lin, Zhang Chen, Zac Yu, Jarvis Jin, Simral Chaudhary, Roman Komarytsia, Christiane Ahlheim, Yonghao Zhu, Bowen Li, Saravanan Ganesh, Bill Byrne, Jessica Hoffmann, Hassan Mansoor, Wei Li, Abhinav Rastogi, Lucas Dixon
  • キーワード: RLHF, Parameter Efficient method, Low Computational Cost, LLM, VLM

• Improving Reinforcement Learning from Human Feedback with Efficient Reward Model Ensemble

  • Shun Zhang, Zhenfang Chen, Sunli Chen, Yikang Shen, Zhiqing Sun, Chuang Gan
  • キーワード: RLHF, Reward Ensemble, Efficient Ensemble Method

• RIME: Robust Preference-based Reinforcement Learning with Noisy Human Preferences

  • Jie Cheng, Gang Xiong, Xingyuan Dai, Qinghai Miao, Yisheng Lv, Fei-Yue Wang
  • キーワード:
  • コード: official

• Uni-RLHF: Universal Platform and Benchmark Suite for Reinforcement Learning with Diverse Human Feedback

  • Yifu Yuan, Jianye Hao, Yi Ma, Zibin Dong, Hebin Liang, Jinyi Liu, Zhixin Feng, Kai Zhao, Yan Zheng
  • キーワード:
  • コード: official
  • データセット: official

2023

• The Trickle-down Impact of Reward (In-)consistency on RLHF

  • Lingfeng Shen, Sihao Chen, Linfeng Song, Lifeng Jin, Baolin Peng, Haitao Mi, Daniel Khashabi, Dong Yu
  • キーワード: Reward model, RLHF, Reward hacking
  • コード: official

• A General Theoretical Paradigm to Understand Learning from Human Preferences

  • Mohammad Gheshlaghi Azar, Mark Rowland, Bilal Piot, Daniel Guo, Daniele Calandriello, Michal Valko, Rémi Munos
  • キーワード: RLHF, Pairwise Preference

• Fine-Grained Human Feedback Gives Better Rewards for Language Model Training

  • Zeqiu Wu, Yushi Hu, Weijia Shi, Nouha Dziri, Alane Suhr, Prithviraj Ammanabrolu, Noah A. Smith, Mari Ostendorf, Hannaneh Hajishirzi
  • キーワード: RLHF, Sentence-level Reward, LLM
  • コード: official

• Preference-grounded Token-level Guidance for Language Model Fine-tuning

  • Shentao Yang, Shujian Zhang, Congying Xia, Yihao Feng, Caiming Xiong, Mingyuan Zhou
  • キーワード: RLHF, Token-level Training Guidance, Alternate/Online Training Framework, Minimalist Training Objectives
  • コード: official

• Fantastic Rewards and How to Tame Them: A Case Study on Reward Learning for Task-oriented Dialogue Systems

  • Yihao Feng*, Shentao Yang*, Shujian Zhang, Jianguo Zhang, Caiming Xiong, Mingyuan Zhou, Huan Wang
  • キーワード: RLHF, Genralized Reward Function Learning, Reward Function Utilization, Task-oriented Dialogue System, Learning-to-rank
  • コード: official

• Inverse Preference Learning: Preference-based RL without a Reward Function

  • Joey Hejna, Dorsa Sadigh
  • キーワード: Inverse Preference Learning, without reward model
  • コード: official

• AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback

  • Yann Dubois, Chen Xuechen Li, Rohan Taori, Tianyi Zhang, Ishaan Gulrajani, Jimmy Ba, Carlos Guestrin, Percy S. Liang, Tatsunori B. Hashimoto
  • キーワード: RLHF, Simulation Framework
  • コード: official

• Adversarial Preference Optimization

  • Pengyu Cheng, Yifan Yang, Jian Li, Yong Dai, Nan Du
  • キーワード: RLHF, GAN, Adversarial Games
  • コード: official

• Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-Constraint

  • Wei Xiong, Hanze Dong, Chenlu Ye, Ziqi Wang, Han Zhong, Heng Ji, Nan Jiang, Tong Zhang
  • キーワード: RLHF, Iterative DPO, Mathematical foundation

• Sample Efficient Reinforcement Learning from Human Feedback via Active Exploration

  • Viraj Mehta, Vikramjeet Das, Ojash Neopane, Yijia Dai, Ilija Bogunovic, Jeff Schneider, Willie Neiswanger
  • キーワード: RLHF, sample efficience, exploration

• Reinforcement Learning from Statistical Feedback: the Journey from AB Testing to ANT Testing

  • Feiyang Han, Yimin Wei, Zhaofeng Liu, Yanxing Qi
  • キーワード: RLHF, AB testing, RLSF

• A Baseline Analysis of Reward Models’ Ability To Accurately Analyze Foundation Models Under Distribution Shift

  • Ben Pikus, Will LeVine, Tony Chen, Sean Hendryx
  • キーワード: RLHF, OOD, Distribution Shift

• Data-Efficient Alignment of Large Language Models with Human Feedback Through Natural Language

  • Di Jin, Shikib Mehri, Devamanyu Hazarika, Aishwarya Padmakumar, Sungjin Lee, Yang Liu, Mahdi Namazifar
  • キーワード: RLHF, data-efficient, Alignment

• Let’s Reinforce Step by Step

  • Sarah Pan, Vladislav Lialin, Sherin Muckatira, Anna Rumshisky
  • キーワード: RLHF, reasoning

• Direct Preference-based Policy Optimization without Reward Modeling

  • Gaon An, Junhyeok Lee, Xingdong Zuo, Norio Kosaka, Kyung-Min Kim, Hyun Oh Song
  • キーワード: RLHF without reward modeling, Contrastive learning, Offline refinforcement learning

• AlignDiff: Aligning Diverse Human Preferences via Behavior-Customisable Diffusion Model

  • Zibin Dong, Yifu Yuan, Jianye Hao, Fei Ni, Yao Mu, Yan Zheng, Yujing Hu, Tangjie Lv, Changjie Fan, Zhipeng Hu
  • キーワード: RLHF, Alignment, Diffusion model

• Eureka: Human-Level Reward Design via Coding Large Language Models

  • Yecheng Jason Ma, William Liang, Guanzhi Wang, De-An Huang, Osbert Bastani, Dinesh Jayaraman, Yuke Zhu, Linxi Fan, Anima Anandkumar
  • キーワード: LLM based, reward functions design

• Safe RLHF: Safe Reinforcement Learning from Human Feedback

  • Josef Dai, Xuehai Pan, Ruiyang Sun, Jiaming Ji, Xinbo Xu, Mickel Liu, Yizhou Wang, Yaodong Yang
  • キーワード: Sale RL, LLM fine-ture

• Quality Diversity through Human Feedback

  • Li Ding, Jenny Zhang, Jeff Clune, Lee Spector, Joel Lehman
  • キーワード: Quality Diversity, Diffusion model

• ReMax: A Simple, Effective, and Efficient Reinforcement Learning Method for Aligning Large Language Models

  • Ziniu Li, Tian Xu, Yushun Zhang, Yang Yu, Ruoyu Sun, Zhi-Quan Luo
  • キーワード: computational efficiency, variance-reduction technique

• Tuning computer vision models with task rewards

  • André Susano Pinto, Alexander Kolesnikov, Yuge Shi, Lucas Beyer, Xiaohua Zhai
  • キーワード: Reward tuning in Computer Vision

• The Wisdom of Hindsight Makes Language Models Better Instruction Followers

  • Tianjun Zhang, Fangchen Liu, Justin Wong, Pieter Abbeel, Joseph E. Gonzalez
  • キーワード: Hindsight Instruction Relabeling, RLHF System, No Value Network Required
  • コード: official

• Language Instructed Reinforcement Learning for Human-AI Coordination

  • Hengyuan Hu, Dorsa Sadigh
  • キーワード: Human-AI coordination, Human preference alignment, Instruction conditioned RL

• Aligning Language Models with Offline Reinforcement Learning from Human Feedback

  • Jian Hu, Li Tao, June Yang, Chandler Zhou
  • キーワード: Decision Transformer-based Alignment, Offline Reinforcement Learning, RLHF System

• Preference Ranking Optimization for Human Alignment

  • Feifan Song, Bowen Yu, Minghao Li, Haiyang Yu, Fei Huang, Yongbin Li and Houfeng Wang
  • キーワード: Supervised Human Preference Alignment, Preference Ranking Extension
  • コード: official

• Bridging the Gap: A Survey on Integrating (Human) Feedback for Natural Language Generation

  • Patrick Fernandes, Aman Madaan, Emmy Liu, António Farinhas, Pedro Henrique Martins, Amanda Bertsch, José G. C. de Souza, Shuyan Zhou, Tongshuang Wu, Graham Neubig, André F. T. Martins
  • キーワード: Natural Language Generation, Human Feedback Integration, Feedback Formalization and Taxonomy, AI Feedback and Principles-Based Judgments

• GPT-4 Technical Report

  • OpenAI
  • キーワード: A large-scale, multimodal model, Transformerbased model, Fine-tuned used RLHF
  • コード: official
  • データセット: DROP, WinoGrande, HellaSwag, ARC, HumanEval, GSM8K, MMLU, TruthfulQA

• RAFT: Reward rAnked FineTuning for Generative Foundation Model Alignment

  • Hanze Dong, Wei Xiong, Deepanshu Goyal, Rui Pan, Shizhe Diao, Jipeng Zhang, Kashun Shum, Tong Zhang
  • キーワード: Rejection Sampling Finetuning, Alternative to PPO, Diffusion Model
  • コード: official

• RRHF: Rank Responses to Align Language Models with Human Feedback without tears

  • Zheng Yuan, Hongyi Yuan, Chuanqi Tan, Wei Wang, Songfang Huang, Fei Huang
  • キーワード: New paradigm for RLHF
  • コード: official

• Few-shot Preference Learning for Human-in-the-Loop RL

  • Joey Hejna, Dorsa Sadigh
  • キーワード: Preference Learning, Interactive Learning, Multi-task Learning, Expanding the pool of available data by viewing human-in-the-loop RL
  • コード: official

• Better Aligning Text-to-Image Models with Human Preference

  • Xiaoshi Wu, Keqiang Sun, Feng Zhu, Rui Zhao, Hongsheng Li
  • キーワード: Diffusion Model, Text-to-Image, Aesthetic
  • コード: official

• ImageReward: Learning and Evaluating Human Preferences for Text-to-Image Generation

  • Jiazheng Xu, Xiao Liu, Yuchen Wu, Yuxuan Tong, Qinkai Li, Ming Ding, Jie Tang, Yuxiao Dong
  • キーワード: General-purpose text-to-Image human preference RM, Evaluating Text-to-Image Generative Models
  • コード: official
  • データセット: COCO, DiffusionDB

• Aligning Text-to-Image Models using Human Feedback

  • Kimin Lee, Hao liu, MoonKyung Ryu, Olivia Watkins, Yuqing Du, Craig Boutilier, Pieter Abbeel, Mohammad Ghavamzadeh, Shixiang Shane Gu
  • キーワード: Text-to-Image, Stable diffusion model, Reward function that predicts human feedback

• Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models

  • Chenfei Wu, Shengming Yin, Weizhen Qi, Xiaodong Wang, Zecheng Tang, Nan Duan
  • キーワード: Visual Foundation Models, Visual ChatGPT
  • コード: official

• Pretraining Language Models with Human Preferences (PHF)

  • Tomasz Korbak, Kejian Shi, Angelica Chen, Rasika Bhalerao, Christopher L. Buckley, Jason Phang, Samuel R. Bowman, Ethan Perez
  • キーワード: Pretraining, offline RL, Decision transformer
  • コード: official

• Aligning Language Models with Preferences through f-divergence Minimization (f-DPG)

  • Dongyoung Go, Tomasz Korbak, Germán Kruszewski, Jos Rozen, Nahyeon Ryu, Marc Dymetman
  • キーワード: f-divergence, RL with KL penalties

• Principled Reinforcement Learning with Human Feedback from Pairwise or K-wise Comparisons

  • Banghua Zhu, Jiantao Jiao, Michael I. Jordan
  • キーワード: Pessimistic MLE, Max-entropy IRL

• The Capacity for Moral Self-Correction in Large Language Models

  • Anthropic
  • キーワード: Improve moral self-correction capability by increasing RLHF training
  • Dataset; BBQ

2022

• Is Reinforcement Learning (Not) for Natural Language Processing?: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization (NLPO)
  • Rajkumar Ramamurthy, Prithviraj Ammanabrolu, Kianté,Brantley, Jack Hessel, Rafet Sifa, Christian Bauckhage, Hannaneh Hajishirzi, Yejin Choi
  • キーワード: Optimizing language generators with RL, Benchmark, Performant RL algorithm
  • コード: official
  • データセット: IMDB, CommonGen, CNN Daily Mail, ToTTo, WMT-16 (en-de),NarrativeQA, DailyDialog
• Scaling Laws for Reward Model Overoptimization
  • Leo Gao, John Schulman, Jacob Hilton
  • キーワード: Gold reward model train proxy reward model, Dataset size, Policy parameter size, BoN, PPO
• Improving alignment of dialogue agents via targeted human judgements (Sparrow)
  • Amelia Glaese, Nat McAleese, Maja Trębacz, et al.
  • キーワード: Information-seeking dialogue agent, Break down the good dialogue into natural language rules, DPC, Interact with the model to elicit violation of a specific rule (Adversarial Probing)
  • データセット: Natural Questions, ELI5, QuALITY, TriviaQA, WinoBias, BBQ
• Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned
  • Deep Ganguli, Liane Lovitt, Jackson Kernion, et al.
  • キーワード: Red team language model, Investigate scaling behaviors, Read teaming Dataset
  • コード: official
• Dynamic Planning in Open-Ended Dialogue using Reinforcement Learning
  • Deborah Cohen, Moonkyung Ryu, Yinlam Chow, Orgad Keller, Ido Greenberg, Avinatan Hassidim, Michael Fink, Yossi Matias, Idan Szpektor, Craig Boutilier, Gal Elidan
  • キーワード: Real-time, Open-ended dialogue system, Pairs the succinct embedding of the conversation state by language models, CAQL, CQL, BERT
• Quark: Controllable Text Generation with Reinforced Unlearning
  • Ximing Lu, Sean Welleck, Jack Hessel, Liwei Jiang, Lianhui Qin, Peter West, Prithviraj Ammanabrolu, Yejin Choi
  • キーワード: Fine-tuning the language model on signals of what not to do, Decision Transformer, LLM tuning with PPO
  • コード: official
  • データセット: WRITINGPROMPTS, SST-2, WIKITEXT-103
• Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
  • Yuntao Bai, Andy Jones, Kamal Ndousse, et al.
  • キーワード: Harmless assistants, Online mode, Robustness of RLHF training, OOD detection.
  • コード: official
  • データセット: TriviaQA, HellaSwag, ARC, OpenBookQA, LAMBADA, HumanEval, MMLU, TruthfulQA
• Teaching language models to support answers with verified quotes (GopherCite)
  • Jacob Menick, Maja Trebacz, Vladimir Mikulik, John Aslanides, Francis Song, Martin Chadwick, Mia Glaese, Susannah Young, Lucy Campbell-Gillingham, Geoffrey Irving, Nat McAleese
  • キーワード: Generate answers which citing specific evidence, Abstain from answering when unsure
  • データセット: Natural Questions, ELI5, QuALITY, TruthfulQA
• Training language models to follow instructions with human feedback (InstructGPT)
  • Long Ouyang, Jeff Wu, Xu Jiang, et al.
  • キーワード: Large Language Model, Align Language Model with Human Intent
  • コード: official
  • データセット: TruthfulQA, RealToxicityPrompts
• Constitutional AI: Harmlessness from AI Feedback
  • Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, et al.
  • キーワード: RL from AI feedback(RLAIF), Training a harmless AI assistant through selfimprovement, Chain-of-thought style, Control AI behavior more precisely
  • コード: official
• Discovering Language Model Behaviors with Model-Written Evaluations
  • Ethan Perez, Sam Ringer, Kamilė Lukošiūtė, Karina Nguyen, Edwin Chen, et al.
  • キーワード: Automatically generate evaluations with LMs, More RLHF makes LMs worse, LM-written evaluations are highquality
  • コード: official
  • データセット: BBQ, Winogender Schemas
• Non-Markovian Reward Modelling from Trajectory Labels via Interpretable Multiple Instance Learning
  • Joseph Early, Tom Bewley, Christine Evers, Sarvapali Ramchurn
  • キーワード: Reward Modelling (RLHF), Non-Markovian, Multiple Instance Learning, Interpretability
  • コード: official
• SURF: Semi-supervised Reward Learning with Data Augmentation for Feedback-efficient Preference-based Reinforcement Learning
  • Jongjin Park, Younggyo Seo, Jinwoo Shin, Honglak Lee, Pieter Abbeel, Kimin Lee
  • キーワード: Semi-supervised Reward Learning, Preference Data Augmentation, RLHF Efficiency
• Reward Uncertainty for Exploration in Preference-based Reinforcement Learning
  • Xinran Liang, Katherine Shu, Kimin Lee, Pieter Abbeel
  • キーワード: Preference-based RL (PbRL), Exploration, Reward Uncertainty, Feedback Efficiency
  • コード: official

2021

• WebGPT: Browser-assisted question-answering with human feedback (WebGPT)
  • Reiichiro Nakano, Jacob Hilton, Suchir Balaji, et al.
  • キーワード: Model search the web and provide reference， Imitation learning， BC, long form question
  • データセット: ELI5, TriviaQA, TruthfulQA
• Recursively Summarizing Books with Human Feedback
  • Jeff Wu, Long Ouyang, Daniel M. Ziegler, Nisan Stiennon, Ryan Lowe, Jan Leike, Paul Christiano
  • キーワード: Model trained on small task to assist human evaluate broader task, BC
  • データセット: Booksum, NarrativeQA
• Revisiting the Weaknesses of Reinforcement Learning for Neural Machine Translation
  • Samuel Kiegeland, Julia Kreutzer
  • キーワード: The success of policy gradient is because of reward rather than the shape of output distribution, Machine Translation, NMT, DOmain Adaption
  • コード: official
  • データセット: WMT15, IWSLT14
• PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training
  • Kimin Lee, Laura Smith, Pieter Abbeel
  • キーワード: Preference-based RL (PbRL), Data Efficiency, Unsupervised Pretraining, Reward Relabeling
  • コード: official
• B-Pref: Benchmarking Preference-Based Reinforcement Learning
  • Kimin Lee, Laura Smith, Anca Dragan, Pieter Abbeel
  • キーワード: Benchmark, Preference-based RL, Simulated Human Feedback, Robustness Evaluation
  • コード: official

2020年以降

• Learning to summarize from human feedback
  • Nisan Stiennon, Long Ouyang, Jeff Wu, Daniel M. Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, Paul Christiano
  • キーワード: Care about summary quality, Training loss affect the model behavior, Reward model generalizes to new datasets
  • コード: official
  • データセット: TL;DR, CNN/DM
• Fine-Tuning Language Models from Human Preferences
  • Daniel M. Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B. Brown, Alec Radford, Dario Amodei, Paul Christiano, Geoffrey Irving
  • キーワード: Reward learning for language, Continuing text with positive sentiment, Summary task, Physical descriptive
  • コード: official
  • データセット: TL;DR, CNN/DM
• Scalable agent alignment via reward modeling: a research direction
  • Jan Leike, David Krueger, Tom Everitt, Miljan Martic, Vishal Maini, Shane Legg
  • キーワード: Agent alignment problem, Learn reward from interaction, Optimize reward with RL, Recursive reward modeling
  • コード: official
  • 環境: Atari
• Reward learning from human preferences and demonstrations in Atari
  • Borja Ibarz, Jan Leike, Tobias Pohlen, Geoffrey Irving, Shane Legg, Dario Amodei
  • キーワード: Expert demonstration trajectory preferences reward hacking problem, Noise in human label
  • コード: official
  • 環境: Atari
• Deep TAMER: Interactive Agent Shaping in High-Dimensional State Spaces
  • Garrett Warnell, Nicholas Waytowich, Vernon Lawhern, Peter Stone
  • キーワード: High dimension state, Leverage the input of Human trainer
  • コード: third party
  • 環境: Atari
• Deep reinforcement learning from human preferences
  • Paul Christiano, Jan Leike, Tom B. Brown, Miljan Martic, Shane Legg, Dario Amodei
  • キーワード: Explore goal defined in human preferences between pairs of trajectories segmentation, Learn more complex thing than human feedback
  • コード: official
  • 環境: Atari, MuJoCo
• Interactive Learning from Policy-Dependent Human Feedback
  • James MacGlashan, Mark K Ho, Robert Loftin, Bei Peng, Guan Wang, David Roberts, Matthew E. Taylor, Michael L. Littman
  • キーワード: Decision is influenced by current policy rather than human feedback, Learn from policy dependent feedback that converges to a local optimal

コードベース

format:
- [title](codebase link) [links]
  - author1, author2, and author3...
  - keyword
  - experiment environments, datasets or tasks

• Reinforcement Learning from Human Feedback (RLHF) in Notebooks
  • Ashwani Kumar
  • step-by-step, Video tutorial, Jupyter notebooks, GPT-2, Reward Model, PPO, Pedagogical
  • データセット: stanfordnlp/sst2
  • タスク: 肯定的な感情でテキストを生成する
  • 環境: Google Colab
• veRL: Volcano Engine Reinforcement Learning for LLM
  • ByteDance Seed MLSys Team & HKU: Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, Chuan Wu
  • キーワード: Flexible, Efficient, RLHF framework
  • タスク: RLHF, 数学と数学を含むタスクを関連付ける code.
• OpenRLHF
  • OpenRLHF
  • キーワード: 70B, RLHF, DeepSpeed, Ray, vLLM
  • タスク: 使いやすい、スケーラブルで高性能 RLHF フレームワーク(サポート70B+フルチューニング&LoRA&Mixtral&KTO)。
• Potato
  • David Jurgens et al.
  • キーワード: Annotation, Human Evaluation, Quality Control, AI-Assisted Labeling
  • タスク: 20以上のアノテーションタイプとエージェントトレース評価による人的評価とフィードバック収集のためのポータブルアノテーションプラットフォーム
• PaLM + RLHF - Pytorch
  • Phil Wang, Yachine Zahidi, Ikko Eltociear Ashimine, Eric Alcaide
  • キーワード: Transformers, PaLM architecture
  • データセット: enwik8
• lm-human-preferences
  • Daniel M. Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B. Brown, Alec Radford, Dario Amodei, Paul Christiano, Geoffrey Irving
  • キーワード: Reward learning for language, Continuing text with positive sentiment, Summary task, Physical descriptive
  • データセット: TL;DR, CNN/DM
• following-instructions-human-feedback
  • Long Ouyang, Jeff Wu, Xu Jiang, et al.
  • キーワード: Large Language Model, Align Language Model with Human Intent
  • データセット: TruthfulQA RealToxicityPrompts
• Transformer Reinforcement Learning (TRL)
  • Leandro von Werra, Younes Belkada, Lewis Tunstall, et al.
  • キーワード: Train LLM with RL, PPO, Transformer
  • タスク: IMDB sentiment
• Transformer Reinforcement Learning X (TRLX)
  • Jonathan Tow, Leandro von Werra, et al.
  • キーワード: Distributed training framework, T5-based language models, Train LLM with RL, PPO, ILQL
  • タスク: ファインチューニング LLM 指定されたリワード機能またはリワードラベルされたデータセットを使用してRL
• RL4LMs (A modular RL library to fine-tune language models to human preferences)
  • Rajkumar Ramamurthy, Prithviraj Ammanabrolu, Kianté,Brantley, Jack Hessel, Rafet Sifa, Christian Bauckhage, Hannaneh Hajishirzi, Yejin Choi
  • キーワード: Optimizing language generators with RL, Benchmark, Performant RL algorithm
  • データセット: IMDB, CommonGen, CNN Daily Mail, ToTTo, WMT-16 (en-de), NarrativeQA, DailyDialog
• LaMDA-rlhf-pytorch
  • Phil Wang
  • キーワード: LaMDA, Attention-mechanism
  • タスク: PyTorch の Google の LaMDA リサーチ ペーパーのオープンソースの事前トレーニングの実施
• TextRL
  • Eric Lam
  • キーワード: huggingface’s transformer
  • タスク: テキスト生成
  • 環境: PFRL, gym
• minRLHF
  • Thomfoster
  • キーワード: PPO, Minimal library
  • タスク: 教育目的
• DeepSpeed-Chat
  • Microsoft
  • キーワード: Affordable RLHF Training
• Dromedary
  • IBM
  • キーワード: Minimal human supervision, Self-aligned
  • タスク: 最小限の人間の監督と訓練される自己一直線の言語モデル
• FG-RLHF
  • Zeqiu Wu, Yushi Hu, Weijia Shi, et al.
  • キーワード: Fine-Grained RLHF, providing a reward after every segment, Incorporating multiple RMs associated with different feedback types
  • タスク: 密度と複数のRMで細かく刻まれた報酬機能のトレーニングと学習を可能にするフレームワーク -Safe-RLHF
  • Xuehai Pan, Ruiyang Sun, Jiaming Ji, et al.
  • キーワード: Support popular pre-trained models, Large human-labeled dataset, Multi-scale metrics for safety constraints verification, Customized parameters
  • タスク: 禁忌の価値-アライメント LLM 安全に RLHF
• VinePPO
  • Amirhossein Kazemnejad, Milad Aghajohari, et al.
  • キーワード: Performant Implementation of RL algorithms for Reasoning, PPO, DPO, RestEM, Monte Carlo Value Estimation
  • タスク: MATHとGSM8Kを含むタスクを関連付ける

データセット

format:
- [title](dataset link) [links]
  - author1, author2, and author3...
  - keyword
  - experiment environments or tasks

• HH-RLHF
  • Ben Mann, Deep Ganguli
  • キーワード: Human preference dataset, Red teaming data, machine-written
  • タスク: 有用性と無害性に関する人間の好みデータのためのオープンソースのデータセット
• Stanford Human Preferences Dataset(SHP)
  • Ethayarajh, Kawin and Zhang, Heidi and Wang, Yizhong and Jurafsky, Dan
  • キーワード: Naturally occurring and human-written dataset,18 different subject areas
  • タスク: 訓練に使用するべき RLHF 報酬モデル
• PromptSource
  • Stephen H. Bach, Victor Sanh, Zheng-Xin Yong et al.
  • キーワード: Prompted English datasets, Mapping a data example into natural language
  • タスク: 自然言語プロンプトの作成、共有、使用するためのツールキット
• Structured Knowledge Grounding(SKG) Resources Collections
  • Tianbao Xie, Chen Henry Wu, Peng Shi et al.
  • キーワード: Structured Knowledge Grounding
  • タスク: データセットの収集は、構造化された知識の接地に関連しています
• The Flan Collection
  • Longpre Shayne, Hou Le, Vu Tu et al.
  • タスク: Flan 2021、P3、Super-Natural Instructions からデータセットをコンパイル
• rlhf-reward-datasets
  • Yiting Xie
  • キーワード: Machine-written dataset
• webgpt_comparisons
  • OpenAI
  • キーワード: Human-written dataset, Long form question answering
  • タスク: 人間の好みと一直線に並ぶモデルに答える長いフォームの質問を訓練して下さい
• summarize_from_feedback
  • OpenAI
  • キーワード: Human-written dataset, summarization
  • タスク: 人間の好みと整列するために要約モデルを訓練
• Dahoas/synthetic-instruct-gptj-pairwise
  • Dahoas
  • キーワード: Human-written dataset, synthetic dataset
• Stable Alignment - Alignment Learning in Social Games
  • Ruibo Liu, Ruixin (Ray) Yang, Qiang Peng
  • キーワード: Interaction data used for alignment training, Run in Sandbox
  • タスク: 記録されたインタラクションデータをシミュレートしたソーシャルゲームでトレイン
• LIMA
  • Meta AI
  • キーワード: without any RLHF, few carefully curated prompts and responses
  • タスク: LIMAモデルのトレーニングに使用されるデータセット

ブログ

• [OpenAI] ChatGPT: Optimizing Language Models for Dialogue
• [Hugging Face] Illustrating Reinforcement Learning from Human Feedback (RLHF)
• [ZhiHu] 通向AGI之路：大型语言模型 (LLM) 技术精要
• [ZhiHu] 大语言模型的涌现能力：现象与解释
• [ZhiHu] 中文hh-rlhf数据集上的ppo实践
• [W&B Fully Connected] Understanding Reinforcement Learning from Human Feedback (RLHF)
• [Deepmind] Learning through human feedback
• [Notion] 深入理解语言模型的突现能力
• [Notion] 拆解追溯 GPT-3.5 各项能力的起源
• [gist] Reinforcement Learning for Language Models
• [YouTube] John Schulman - Reinforcement Learning from Human Feedback: Progress and Challenges
• [OpenAI / Arize] OpenAI on Reinforcement Learning With Human Feedback
• [Encord] Guide to Reinforcement Learning from Human Feedback (RLHF) for Computer Vision
• [hijkzzz] A Survey of Reinforcement Learning from Human Feedback (RLHF)
• [Weixun Wang] Overview of RL(HF)+LLM
• [Lilian Weng] Reward Hacking in Reinforcement Learning

出版書籍

• Reinforcement Learning from Human Feedback by Nathan Lambert
• Reinforcement Learning for Business
• The RLHF Book

その他の言語サポート

Turkish

貢献する

私たちの目的は、このレポをより良いものにすることです。 ご協力のほどよろしくお願いします。 HERE 貢献の指示のため。

ライセンス

アウィサム RLHF Apache 2.0 ライセンスでリリースされます。