ISM:Self-Improving Strategy Memory for Continual Mathematical Reasoning
For LLMs in continual learning settings, ISM provides a parameter-free method to maintain reliable reasoning with compact, verified strategy memories.
ISM is a self-evolving memory system that improves mathematical reasoning in frozen LLMs under continual learning with hard resets, outperforming baselines on MATH-Hard and OlympiadBench while using 64% and 86% fewer schemas respectively.
We propose Intelligent Schema Memory (ISM), a self-evolving memory-augmented system that improves mathematical reasoning for a frozen LLM under continual learning with hard episodic resets. ISM maintains a compact, self-refined bank of strategy schemas learned from both successful and failed episodes, with symbolic tools that check intermediate steps and certify answers.Without updating model parameters, ISM outperforms passive, retrieval, and reflection baselines on MATH-Hard and OlympiadBench, using 64% and 86% fewer schemas respectively than the strongest passive baseline. These results show that small, actively maintained, and verified strategy memories can support reliable continual mathematical reasoning under strict episodic isolation.