From Memory to Skills: Evidence-Grounded Co-Evolution Governance for Long-Horizon LLM Agents

Short summary

MSCE is a training-free framework that converts LLM agent experience traces into reusable, callable skills with evidence links and reliability estimates. It uses reflection-weighted value backfilling to propagate sparse terminal feedback into dense local self-reflections, calibrating trace values for governing memory and skill evolution. Experiments on EvoAgentBench and LoCoMo show MSCE outperforms state-of-the-art skill-augmented and memory-driven baselines with strong cross-domain transferability.

• •Proposes MSCE: a training-free Memory-Skill Co-Evolution framework for long-horizon LLM agents
• •Crystallizes evidence-backed policies into callable skills with verification rules and reliability estimates
• •Outperforms SOTA baselines on EvoAgentBench and LoCoMo with strong cross-domain transfer

Generated with AI, which can make mistakes.