迈向多模态大语言模型的元认知知识编辑
Towards Meta-Cognitive Knowledge Editing for Multimodal LLMs
AI总结：
本文提出CogEdit基准和MIND框架，从三个层面评估并提升多模态大语言模型的元认知知识编辑能力，实验证明MIND显著优于现有方法。
摘要：
知识编辑使多模态大语言模型（MLLMs）能够高效地更新过时或错误的信息。然而，现有基准主要强调认知层面的修改，而缺乏对更深层元认知过程的关注。为弥补这一差距，我们引入了CogEdit，这是一个新颖的基准，旨在从三个层面评估MLLMs的元认知知识编辑能力：（1）反事实驱动编辑，评估对知识正确性变化的自我意识；（2）边界约束编辑，确保在无意外干扰的情况下进行适当泛化；（3）噪声鲁棒编辑，促进对不确定信息的反思性评估。为推进元认知编辑，我们提出了MIND（元认知集成动态知识编辑），这是一个构建元知识记忆以实现自我意识、采用博弈论交互来监控知识激活、并结合标签细化以实现噪声鲁棒更新的框架。大量实验表明，MIND显著优于现有的认知编辑方法，在传统和元认知知识编辑基准上均取得了强劲性能。
英文摘要：
Knowledge editing enables multimodal large language models (MLLMs) to efficiently update outdated or incorrect information. However, existing benchmarks primarily emphasize cognitive-level modifications while lacking a focus on deeper meta-cognitive processes. To bridge this gap, we introduce CogEdit, a novel benchmark designed to evaluate MLLMs' meta-cognitive knowledge editing abilities across three levels: (1) Counterfactual-Driven Editing, assessing self-awareness of knowledge correctness changes; (2) Boundary Constraint Editing, ensuring appropriate generalization without unintended interference; and (3) Noise-Robust Editing, promoting reflective evaluation of uncertain information. To advance meta-cognitive editing, we propose MIND (Meta-cognitive INtegrated Dynamic Knowledge Editing), a framework that constructs a meta-knowledge memory for self-awareness, employs game-theoretic interactions to monitor knowledge activation, and incorporates label refinement for noise-robust updates. Extensive experiments show that MIND significantly outperforms existing cognitive editing approaches, achieving strong performance on both traditional and meta-cognitive knowledge editing benchmarks.
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。
