arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2604.17399cs.AI
收藏

超越元推理:面向自我改进大语言模型推理的元认知巩固

Beyond Meta-Reasoning: Metacognitive Consolidation for Self-Improving LLM Reasoning

Southeast University(东南大学);King’s College London(伦敦国王学院);The Alan Turing Institute(艾伦·图灵研究所);Qilu University of Technology (Shandong Academy of Sciences)(齐鲁工业大学(山东科学院))

Ziqing Zhuang, Linhai Zhang, Jiasheng Si, Deyu Zhou, Yulan He

AI总结:

本文提出元认知巩固框架,通过整合历史推理经验提升未来元推理能力,实验显示性能随经验积累而提升。

摘要:

大型语言模型(LLMs)展现了强大的推理能力,但现有元推理方法多为片段化,缺乏跨实例的知识积累。本文引入元认知巩固框架,将实例级问题解决划分为推理、监控与控制角色,生成可追溯的元级轨迹,并通过分层多时间尺度更新机制逐步形成演化的元知识。实验结果表明,该方法在多个基准和基础模型上均取得一致性能提升,且性能随元认知经验的积累而持续改善。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑