Beyond Meta-Reasoning: Metacognitive Consolidation for Self-Improving LLM Reasoning
For LLM reasoning research, this work addresses the limitation of episodic meta-reasoning by enabling cross-instance skill accumulation, but the gains are incremental over existing methods.
The paper introduces Metacognitive Consolidation, a framework that enables LLMs to accumulate and reuse meta-reasoning skills across instances, leading to consistent performance gains across benchmarks and backbone models, with performance improving as metacognitive experience accumulates.
Large language models (LLMs) have demonstrated strong reasoning capabilities, and as existing approaches for enhancing LLM reasoning continue to mature, increasing attention has shifted toward meta-reasoning as a promising direction for further improvement. However, most existing meta-reasoning methods remain episodic: they focus on executing complex meta-reasoning routines within individual instances, but ignore the accumulation of reusable meta-reasoning skills across instances, leading to recurring failure modes and repeatedly high metacognitive effort. In this paper, we introduce Metacognitive Consolidation, a novel framework in which a model consolidates metacognitive experience from past reasoning episodes into reusable knowledge that improves future meta-reasoning. We instantiate this framework by structuring instance-level problem solving into distinct roles for reasoning, monitoring, and control to generate rich, attributable meta-level traces. These traces are then consolidated through a hierarchical, multi-timescale update mechanism that gradually forms evolving meta-knowledge. Experimental results demonstrate consistent performance gains across benchmarks and backbone models, and show that performance improves as metacognitive experience accumulates over time.