Time Sensitive Knowledge Editing through Efficient Finetuning

TL;DR

This paper introduces a parameter-efficient fine-tuning approach for updating and injecting knowledge into large language models, addressing limitations of existing methods in multi-hop reasoning and scalability.

Contribution

It proposes PEFT for knowledge editing, creates a comprehensive temporal KE dataset, and demonstrates PEFT's advantages over locate-and-edit methods.

Findings

PEFT outperforms locate-and-edit methods in time-sensitive knowledge editing.

Fine-tuning specific layers enhances multi-hop question answering.

The new dataset benchmarks knowledge editing performance across various scenarios.

Abstract

Large Language Models (LLMs) have demonstrated impressive capability in different tasks and are bringing transformative changes to many domains. However, keeping the knowledge in LLMs up-to-date remains a challenge once pretraining is complete. It is thus essential to design effective methods to both update obsolete knowledge and induce new knowledge into LLMs. Existing locate-and-edit knowledge editing (KE) method suffers from two limitations. First, the post-edit LLMs by such methods generally have poor capability in answering complex queries that require multi-hop reasoning. Second, the long run-time of such locate-and-edit methods to perform knowledge edits make it infeasible for large scale KE in practice. In this paper, we explore Parameter-Efficient Fine-Tuning (PEFT) techniques as an alternative for KE. We curate a more comprehensive temporal KE dataset with both knowledge update and knowledge injection examples for KE performance benchmarking. We further probe the effect of fine-tuning on a range of layers in an LLM for the multi-hop QA task. We find that PEFT performs better than locate-and-edit techniques for time-sensitive knowledge edits.