Search

Few direct matches — filled in with the latest updates.

Tag: #llm-rl1 results

MEL Boosts LLM Reasoning via Meta-Experience

MEL Boosts LLM Reasoning via Meta-Experience

Meta-Experience Learning (MEL) enhances RLVR by internalizing error-derived meta-experience into LLM memory. Uses self-verification for contrastive analysis of trajectories. Achieves 3.92%-4.73% Pass@1 gains across model sizes.

ArXiv AIResearchFeb 12#research#mel#v1