提出MGV框架，让大模型像人一样在推理前评估难易、判断信心，避免过早陷入错误路径。
Monitor-Generate-Verify (MGV): Formalising Metacognitive Theory for Language Model Reasoning
- 引入监控模块，在生成前评估任务难度与自身信心
- 通过验证反馈动态优化监控判断，减少20%的准确率损失风险
- 为推理系统诊断和未来设计提供心理学依据，适合研究认知机制者
测试时推理架构如生成-验证范式虽强调生成与验证，却忽略了决定何时及如何启动推理的监控过程。这一缺失可能导致前缀主导陷阱：模型过早锁定次优推理路径且难以修正，造成约20%的准确率下降。本文提出监测-生成-验证（MGV）框架，基于Flavell及Nelson与Narens的元认知理论构建计算模型，保留其心理细节。MGV在生成前加入显式监控，捕捉从难度评估到信心判断的元认知体验，并通过验证反馈不断优化后续监控。尽管无实证验证，但MGV为推理系统组件级故障诊断提供术语体系，提出具体架构改进方向，并揭示与资源理性分析的联系，或可使机制建立在规范性原则之上。
原文摘要 · Abstract (English)
Test-time reasoning architectures such as those following the Generate-Verify paradigm, where a model iteratively refines or verifies its own generated outputs, prioritise generation and verification but exclude the monitoring processes that determine when and how reasoning should begin. This omission may contribute to the prefix dominance trap, in which models commit early to suboptimal reasoning paths and seldom recover, yielding roughly 20% accuracy loss. We address this architectural gap by proposing the Monitor-Generate-Verify (MGV) framework, a computational translation of Flavell's and Nelson and Narens' metacognitive theories that preserves their psychological detail. MGV extends the Generate-Verify paradigm by adding explicit monitoring that captures metacognitive experiences (from difficulty assessments to confidence judgements) before generation begins and refines future monitoring through verification feedback. Though we present no empirical validation, MGV provides a vocabulary for diagnosing component-level failures in reasoning systems, suggests specific architectural interventions for future designs, and identifies connections to resource-rational analysis that may ground its mechanisms in normative principles.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。