0

Safe RLHF: Safe Reinforcement Learning from Human Feedback

Safe RLHF, a novel algorithm for human value alignment in LLMs, improves performance and safety by decoupling human preferences and using a Lagrangian method to balance reward and cost constraints.

Year
2023
Venue
arXiv 2023
Authors
8
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2310.12773ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

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 us to train separate reward and cost models. We formalize the safety concern of LLMs as an optimization task of maximizing the reward function while satisfying specified cost constraints. Leveraging the Lagrangian method to solve this constrained problem, Safe RLHF dynamically adjusts the balance between the two objectives during fine-tuning. Through a three-round fine-tuning using Safe RLHF, we demonstrate a superior ability to mitigate harmful responses while enhancing model performance compared to existing value-aligned algorithms. Experimentally, we fine-tuned the Alpaca-7B using Safe RLHF and aligned it with collected human preferences, significantly improving its helpfulness and harmlessness according to human evaluations.

Authors

8