大型语言模型知识编辑的全面研究

摘要

大型语言模型（LLMs）展现出在理解和生成文本方面的非凡能力，其文本与人类交流密切相似。然而，其主要限制在于训练过程中存在的巨大计算需求，这是由于其广泛的参数化所导致的。这一挑战进一步加剧了世界动态性的影响，需要经常更新LLMs以纠正过时信息或整合新知识，从而确保其持续相关性。需要指出的是，许多应用需要在训练后持续调整模型以解决缺陷或不良行为。对于即时模型修改，高效轻量级方法的兴趣日益增加。为此，近年来知识编辑技术呈现蓬勃发展，旨在有效修改LLMs在特定领域内的行为，同时保持其在各种输入下的整体性能。在本文中，我们首先定义知识编辑问题，然后全面审视最前沿的方法。借鉴教育和认知研究理论，我们提出一个统一的分类标准，将知识编辑方法分为三类：利用外部知识、将知识合并到模型中以及编辑内在知识。此外，我们引入了一个新的基准，KnowEdit，用于全面实证评估代表性的知识编辑方法。此外，我们对知识定位进行了深入分析，这可以更深入地理解LLMs内在的知识结构。最后，我们讨论了知识编辑的几个潜在应用，概述了其广泛而深远的影响。

English

Large Language Models (LLMs) have shown extraordinary capabilities in understanding and generating text that closely mirrors human communication. However, a primary limitation lies in the significant computational demands during training, arising from their extensive parameterization. This challenge is further intensified by the dynamic nature of the world, necessitating frequent updates to LLMs to correct outdated information or integrate new knowledge, thereby ensuring their continued relevance. Note that many applications demand continual model adjustments post-training to address deficiencies or undesirable behaviors. There is an increasing interest in efficient, lightweight methods for on-the-fly model modifications. To this end, recent years have seen a burgeoning in the techniques of knowledge editing for LLMs, which aim to efficiently modify LLMs' behaviors within specific domains while preserving overall performance across various inputs. In this paper, we first define the knowledge editing problem and then provide a comprehensive review of cutting-edge approaches. Drawing inspiration from educational and cognitive research theories, we propose a unified categorization criterion that classifies knowledge editing methods into three groups: resorting to external knowledge, merging knowledge into the model, and editing intrinsic knowledge. Furthermore, we introduce a new benchmark, KnowEdit, for a comprehensive empirical evaluation of representative knowledge editing approaches. Additionally, we provide an in-depth analysis of knowledge location, which can provide a deeper understanding of the knowledge structures inherent within LLMs. Finally, we discuss several potential applications of knowledge editing, outlining its broad and impactful implications.

大型语言模型知识编辑的全面研究

A Comprehensive Study of Knowledge Editing for Large Language Models

摘要

Support