ChainEdit: Propagating Ripple Effects in LLM Knowledge Editing through Logical Rule-Guided Chains

Current knowledge editing methods for large language models (LLMs) struggle to maintain logical consistency when propagating ripple effects to associated facts. We propose ChainEdit, a framework that synergizes knowledge graph-derived logical rules with LLM logical reasoning capabilities to enable systematic chain updates. By automatically extracting logical patterns from structured knowledge bases and aligning them with LLMs' internal logics, ChainEdit dynamically generates and edits logically connected knowledge clusters. Experiments demonstrate an improvement of more than 30% in logical generalization over baselines while preserving editing reliability and specificity. We further address evaluation biases in existing benchmarks through knowledge-aware protocols that disentangle external dependencies. This work establishes new state-of-the-art performance on ripple effect while ensuring internal logical consistency after knowledge editing. The code will be available at https://github.com/NUSTM/ChainEdit . Recent years have witnessed a continuous expansion of large language models' (LLMs) capabilities along with increasing model parameters. As new information constantly emerges and existing knowledge evolves, it is crucial to keep LLMs up-to-date and accurate. However, retraining these models to reflect such changes is prohibitively expensive and time-consuming due to their massive parameter sizes. This highlights the importance of knowledge editing techniques for LLMs, which allow for targeted modifications without the need for full retraining, thus offering an efficient alternative to traditional methods. Knowledge editing approaches can be categorized into parameter-preserving and parameter-* Equal Contribution. † Corresponding Author.