Generalizable Multimodal Large Language Model Editing via Invariant Trajectory Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917234354421760 |
|---|---|
| author | Su, Jiajie Wang, Haoyuan Feng, Xiaohua Ma, Yunshan Xia, Xiaobo Li, Yuyuan Zheng, Xiaolin Xiao, Jianmao Chen, Chaochao |
| author_facet | Su, Jiajie Wang, Haoyuan Feng, Xiaohua Ma, Yunshan Xia, Xiaobo Li, Yuyuan Zheng, Xiaolin Xiao, Jianmao Chen, Chaochao |
| contents | Knowledge editing emerges as a crucial technique for efficiently correcting incorrect or outdated knowledge in large language models (LLM).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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_19700 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Generalizable Multimodal Large Language Model Editing via Invariant Trajectory Learning Su, Jiajie Wang, Haoyuan Feng, Xiaohua Ma, Yunshan Xia, Xiaobo Li, Yuyuan Zheng, Xiaolin Xiao, Jianmao Chen, Chaochao Machine Learning Artificial Intelligence Knowledge editing emerges as a crucial technique for efficiently correcting incorrect or outdated knowledge in large language models (LLM).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. |
| title | Generalizable Multimodal Large Language Model Editing via Invariant Trajectory Learning |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2601.19700 |