arXiv:2605.28860v2 Announce Type: replace-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 citeshenfeld2025rl. We extend this behavioral account to the mechanistic level and ask whether RL’s advantage is mirrored by stronger preservation of internal computational circuits. We introduce differential circuit vulnerability, a head-level measure of how much a circuit degrades under fine-tuning, and use it to compare RL and SFT on Qwen2.5-3B-Instruct adapted to scientific question-answering. We find a clear mechanistic trade-off: SFT adapts more rapidly to the target task but produces substantially greater circuit disruption and forgetting of prior capabilities, whereas RL preserves a larger fraction of the base circuit at the cost of slower task adaptation. These findings suggest that circuit preservation may help explain why RL is more robust to catastrophic forgetting. We released our code here: https://github.com/rl-sft-circuit-research/differential-circuit-vulnerability.
Crisis support teams’ technological openness and learning attitudes toward the AI based virtual patient system crisis support VR
BackgroundAgainst the backdrop of escalating global humanitarian crises, innovative didactic simulations are becoming increasingly important. A promising alternative to traditional classroom-based didactics for learning psychological