Mechanistic origins of catastrophic forgetting: why RL preserves circuits better than SFT? 事件

Mechanistic origins of catastrophic forgetting: why RL preserves circuits better than SFT? arXiv:2605.28860v1 Announce Type: cross Abstract: Fine-tuning large language models (LLMs) frequently induces catastrophic forgetting of prior capabilities. Recent work has shown that reinforcement learning (RL) retains prior capabilities more effectively than supervised fine-tuning (SFT), attributing this to policy-gradient updates remaining closer to the base policy \cite{shenfeld2025rl}. We extend this