Title:Reviving Dormant Memories: Investigating Catastrophic Forgetting in Language Models through Rationale-Guidance Difficulty

Abstract:Although substantial efforts have been made to mitigate catastrophic forgetting in continual learning, the intrinsic mechanisms are not well understood. In this paper, we discover that when a forgetting model passively receives an externally provided partial appropriate rationale, its performance on the forgotten task can be restored. Furthermore, by simply adding a task-agnostic prefix to the original instruction, the forgetting model can actively generate an appropriate rationale to reach the correct answer. These findings suggest that the model does not actually ``forget'' the task knowledge; instead, the degraded performance can be attributed to the failure of the original instructions in guiding the model to generate the appropriate rationales. Based on this insight, we propose the Rationale-Guidance Difficulty metric to evaluate how effectively a given instruction guides the model in generating appropriate rationales. We apply this metric to optimize the allocation of replay data in replay-based continual learning algorithm. Experimental results demonstrate that our data allocation method effectively mitigates catastrophic forgetting and maintains better model plasticity simultaneously across models.