RLHF Workflow: From Reward Modeling to Online RLHF
We present the workflow of Online Iterative Reinforcement Learning from Human Feedback (RLHF) in this technical report, which is widely reported to outperform its offline counterpart by a large margin in the recent large language model (LLM) literature. However, existing open-source RLHF projects are still largely confined to the offline learning setting. In this technical report, we aim to fill in this gap and provide a detailed recipe that is easy to reproduce for online iterative RLHF. In particular, since online human feedback is usually infeasible for open-source communities with limited resources, we start by constructing preference models using a diverse set of open-source datasets and use the constructed proxy preference model to approximate human feedback. Then, we discuss the theoretical insights and algorithmic principles behind online iterative RLHF, followed by a detailed practical implementation. Our trained LLM achieves impressive performance on LLM chatbot benchmarks, including AlpacaEval-2, Arena-Hard, and MT-Bench, as well as other academic benchmarks such as HumanEval and TruthfulQA. We have shown that supervised fine-tuning (SFT) and iterative RLHF can obtain state-of-the-art performance with fully open-source datasets. Further, we have made our models, curated datasets, and comprehensive step-by-step code guidebooks publicly available. Please refer to https://github.com/RLHFlow/RLHF-Reward-Modeling and https://github.com/RLHFlow/Online-RLHF for more detailed information.
Code (3)
Tasks
ChatbotHumanEvalLanguage ModellingLarge Language ModelTruthfulQAMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Value-Incentivized Preference Optimization: A Unified Approach to Online and Offline RLHF
Reinforcement learning from human feedback (RLHF) has demonstrated great promise in aligning large language models (LLMs) with human preference. Depending on the availability of preference data, both online and offline R…
reinforcement-learningReinforcement LearningReinforcement Learning (RL)Text SummarizationPolicy-labeled Preference Learning: Is Preference Enough for RLHF?
To design rewards that align with human goals, Reinforcement Learning from Human Feedback (RLHF) has emerged as a prominent technique for learning reward functions from human preferences and optimizing policies via reinf…
continuous-controlContinuous ControlDecision Makingreinforcement-learning+2G-Core: A Simple, Scalable and Balanced RLHF Trainer
Reinforcement Learning from Human Feedback (RLHF) has become an increasingly popular paradigm for training large language models (LLMs) and diffusion models. While existing RLHF training systems have enabled significant …
Reinforcement LearningEnhancing RLHF with Human Gaze Modeling
Reinforcement Learning from Human Feedback (RLHF) aligns language models with human preferences but is computationally expensive. We explore two approaches that leverage human gaze modeling to enhance RLHF: (1) gaze-awar…
Reinforcement LearningAvoiding $\mathbf{exp(R_{max})}$ scaling in RLHF through Preference-based Exploration
Reinforcement Learning from Human Feedback (RLHF) has emerged as a pivotal technique for large language model (LLM) alignment. This paper studies the setting of online RLHF and focus on improving sample efficiency. All e…
Language ModelingLanguage ModellingLarge Language Model