Model Merging for Knowledge Editing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fu, Zichuan, Wu, Xian, Li, Guojing, Zhang, Yingying, Zheng, Yefeng, Ming, Tianshi, Wang, Yejing, Wang, Wanyu, Zhao, Xiangyu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909649147527168
author Fu, Zichuan
Wu, Xian
Li, Guojing
Zhang, Yingying
Zheng, Yefeng
Ming, Tianshi
Wang, Yejing
Wang, Wanyu
Zhao, Xiangyu
author_facet Fu, Zichuan
Wu, Xian
Li, Guojing
Zhang, Yingying
Zheng, Yefeng
Ming, Tianshi
Wang, Yejing
Wang, Wanyu
Zhao, Xiangyu
contents Large Language Models (LLMs) require continuous updates to maintain accurate and current knowledge as the world evolves. While existing knowledge editing approaches offer various solutions for knowledge updating, they often struggle with sequential editing scenarios and harm the general capabilities of the model, thereby significantly hampering their practical applicability. This paper proposes a two-stage framework combining robust supervised fine-tuning (R-SFT) with model merging for knowledge editing. Our method first fine-tunes the LLM to internalize new knowledge fully, then merges the fine-tuned model with the original foundation model to preserve newly acquired knowledge and general capabilities. Experimental results demonstrate that our approach significantly outperforms existing methods in sequential editing while better preserving the original performance of the model, all without requiring any architectural changes. Code is available at: https://github.com/Applied-Machine-Learning-Lab/MM4KE.
format Preprint
id arxiv_https___arxiv_org_abs_2506_12384
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Model Merging for Knowledge Editing
Fu, Zichuan
Wu, Xian
Li, Guojing
Zhang, Yingying
Zheng, Yefeng
Ming, Tianshi
Wang, Yejing
Wang, Wanyu
Zhao, Xiangyu
Artificial Intelligence
Computation and Language
Machine Learning
Large Language Models (LLMs) require continuous updates to maintain accurate and current knowledge as the world evolves. While existing knowledge editing approaches offer various solutions for knowledge updating, they often struggle with sequential editing scenarios and harm the general capabilities of the model, thereby significantly hampering their practical applicability. This paper proposes a two-stage framework combining robust supervised fine-tuning (R-SFT) with model merging for knowledge editing. Our method first fine-tunes the LLM to internalize new knowledge fully, then merges the fine-tuned model with the original foundation model to preserve newly acquired knowledge and general capabilities. Experimental results demonstrate that our approach significantly outperforms existing methods in sequential editing while better preserving the original performance of the model, all without requiring any architectural changes. Code is available at: https://github.com/Applied-Machine-Learning-Lab/MM4KE.
title Model Merging for Knowledge Editing
topic Artificial Intelligence
Computation and Language
Machine Learning
url https://arxiv.org/abs/2506.12384