Knowledge editing emerges as a crucial technique for efficiently correcting incorrect or outdated knowledge in large language models (LLM). Existing editing methods rely on a rigid mapping from parameter or module modifications to output, which causes the generalization limitation in Multimodal LLM (MLLM). In this paper, we reformulate MLLM editing as an out-of-distribution (OOD) generalization problem, where the goal is to discern semantic shift with factual shift and thus achieve robust editing among diverse cross-modal prompting. The key challenge of this OOD problem lies in identifying invariant causal trajectories that generalize accurately while suppressing spurious correlations. To address it, we propose ODEdit, a plug-and-play invariant learning based framework that optimizes the tripartite OOD risk objective to simultaneously enhance editing reliability, locality, and generality.We further introduce an edit trajectory invariant learning method, which integrates a total variation penalty into the risk minimization objective to stabilize edit trajectories against environmental variations. Theoretical analysis and extensive experiments demonstrate the effectiveness of ODEdit.
@article{arxiv.2601.19700,
title = {Generalizable Multimodal Large Language Model Editing via Invariant Trajectory Learning},
author = {Jiajie Su and Haoyuan Wang and Xiaohua Feng and Yunshan Ma and Xiaobo Xia and Yuyuan Li and Xiaolin Zheng and Jianmao Xiao and Chaochao Chen},
journal= {arXiv preprint arXiv:2601.19700},
year = {2026}
}