让大模型更懂常识，通过动态诊断和概念扩展提升推理能力
ConKE: Conceptualization-Augmented Knowledge Editing in Large Language Models for Commonsense Reasoning
- 用另一大模型动态检测并修正常识知识错误
- 通过概念化扩展知识，使修改后的内容泛化能力更强
- 适合需要提升常识推理的模型优化与应用开发
知识编辑（KE）旨在调整大语言模型（LLM）的内部表征和参数，以纠正不准确信息并提升输出一致性，而无需重新训练整个模型。然而，现有方法在处理常识知识时仍面临知识覆盖有限、标注海量常识不现实、以及格式要求严格等问题。本文提出 ConceptEdit 框架，将概念化与实例化融入 KE 流程，增强 LLM 的常识推理能力。该框架利用另一个验证型 LLM 动态诊断模型中的不合理常识，并对需编辑的知识进行概念化扩展，以增强泛化性。实验表明，采用 ConceptEdit 后的 LLM 在生成常识内容的合理性上优于其他基线，在多个问答基准测试中表现更优。数据、代码与模型已公开于 https://github.com/HKUST-KnowComp/ConKE。
原文摘要 · Abstract (English)
Knowledge Editing (KE) aims to adjust a Large Language Model's (LLM) internal representations and parameters to correct inaccuracies and improve output consistency without incurring the computational expense of re-training the entire model. However, editing commonsense knowledge still faces difficulties, including limited knowledge coverage in existing resources, the infeasibility of annotating labels for an overabundance of commonsense knowledge, and the strict knowledge formats of current editing methods. In this paper, we address these challenges by presenting ConceptEdit, a framework that integrates conceptualization and instantiation into the KE pipeline for LLMs to enhance their commonsense reasoning capabilities. ConceptEdit dynamically diagnoses implausible commonsense knowledge within an LLM using another verifier LLM and augments the source knowledge to be edited with conceptualization for stronger generalizability. Experimental results demonstrate that LLMs enhanced with ConceptEdit successfully generate commonsense knowledge with improved plausibility compared to other baselines and achieve stronger performance across multiple question answering benchmarks. Our data, code, and models are publicly available at https://github.com/HKUST-KnowComp/ConKE.
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。