Safe RLHF: Safe Reinforcement Learning from Human Feedback

Josef Dai, Xuehai Pan, Ruiyang Sun et al.

2023 · arXiv (Cornell University) · 20 citations

With the development of large language models (LLMs), striking a balance between the performance and safety of AI systems has never been more critical. However, the inherent tension between the objectives of helpfulness and harmlessness presents a significant challenge during LLM training. To address this issue, we propose Safe Reinforcement Learning from Human Feedback (Safe RLHF), a novel algorithm for human value alignment. Safe RLHF explicitly decouples human preferences regarding helpfulness and harmlessness, effectively avoiding the crowdworkers' confusion about the tension and allowing…

Read the paper →

Explore this paper's citation graph on Constellation.