Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback

Abstract

Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there has been relatively little public work systematizing its flaws. In this paper, we (1) survey open problems and fundamental limitations of RLHF and related methods; (2) overview techniques to understand, improve, and complement RLHF in practice; and (3) propose auditing and disclosure standards to improve societal oversight of RLHF systems. Our work emphasizes the limitations of RLHF and highlights the importance of a multi-faceted approach to the development of safer AI systems.

Cited by in corpus (10)

• A Comprehensive Survey of Large Language Models and Multimodal Large Language Models in Medicine
• UAVs Meet LLMs: Overviews and Perspectives Toward Agentic Low-Altitude Mobility
• Black-Box Access is Insufficient for Rigorous AI Audits
• Reinforcement Learning for Generative AI: State of the Art, Opportunities and Open Research Challenges
• Beyond Preferences in AI Alignment
• Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Duties
• Reinforcement Learning from Human Feedback: Whose Culture, Whose Values, Whose Perspectives?
• Bootstrapping Cognitive Agents with a Large Language Model
• Generative AI in Health Economics and Outcomes Research: A Taxonomy of Key Definitions and Emerging Applications, an ISPOR Working Group Report
• FARPLS: A Feature-Augmented Robot Trajectory Preference Labeling System to Assist Human Labelers' Preference Elicitation