DAFNet: Dynamic Auxiliary Fusion for Sequential Model Editing in Large Language Models

TL;DR

This paper introduces DAFNet, a novel model for sequentially editing large language models to correct factual errors continuously, using dynamic fusion techniques to prevent forgetting and improve editing accuracy.

Contribution

The paper proposes DAFNet with a new semantic fusion mechanism and introduces DAFSet, a dataset designed for robust sequential model editing in large language models.

Findings

DAFNet outperforms existing baselines in single-turn and sequential editing tasks.

The use of DAFSet enhances the performance of auxiliary network-based editing methods.

DAFNet effectively prevents catastrophic forgetting during multiple knowledge updates.

Abstract

Recently, while large language models (LLMs) have demonstrated impressive results, they still suffer from hallucination, i.e., the generation of false information. Model editing is the task of fixing factual mistakes in LLMs; yet, most previous works treat it as a one-time task, paying little attention to ever-emerging mistakes generated by LLMs. We address the task of sequential model editing (SME) that aims to rectify mistakes continuously. A Dynamic Auxiliary Fusion Network (DAFNet) is designed to enhance the semantic interaction among the factual knowledge within the entire sequence, preventing catastrophic forgetting during the editing process of multiple knowledge triples. Specifically, (1) for semantic fusion within a relation triple, we aggregate the intra-editing attention flow into auto-regressive self-attention with token-level granularity in LLMs. We further leverage multi-layer diagonal inter-editing attention flow to update the weighted representations of the entire sequence-level granularity. (2) Considering that auxiliary parameters are required to store the knowledge for sequential editing, we construct a new dataset named \textbf{DAFSet}, fulfilling recent, popular, long-tail and robust properties to enhance the generality of sequential editing. Experiments show DAFNet significantly outperforms strong baselines in single-turn and sequential editing. The usage of DAFSet also consistently improves the performance of other auxiliary network-based methods in various scenarios