TY - GEN
T1 - M2Edit
T2 - 30th Conference on Empirical Methods in Natural Language Processing, EMNLP 2025
AU - Zhou, Yang
AU - Cao, Pengfei
AU - Chen, Yubo
AU - Liu, Qingbin
AU - Sui, Dianbo
AU - Chen, Xi
AU - Liu, Kang
AU - Zhao, Jun
N1 - Publisher Copyright:
© 2025 Association for Computational Linguistics.
PY - 2025
Y1 - 2025
N2 - Multimodal knowledge editing is an important method for modifying outdated or incorrect knowledge in Multimodal Large Language Models (MLLMs). However, existing datasets for multimodal knowledge editing lack multi-granularity knowledge. In this paper, we present a more realistic dataset called M2Edit, which includes three distinct types of knowledge: entity, relation, and action. Additionally, existing knowledge editing methods for MLLMs lack the ability to handle multi-granularity knowledge and generalize to multimodal data. To address these limitations, we propose the multimodal knowledge editing method MLE. This approach identifies key knowledge layers within different components and collaboratively edits the various components of MLLMs. As a result, we observe significant improvements in visual generality performance, ranging from 4.8% to 10.8%, and achieve the best overall performance on knowledge data of different granularities.
AB - Multimodal knowledge editing is an important method for modifying outdated or incorrect knowledge in Multimodal Large Language Models (MLLMs). However, existing datasets for multimodal knowledge editing lack multi-granularity knowledge. In this paper, we present a more realistic dataset called M2Edit, which includes three distinct types of knowledge: entity, relation, and action. Additionally, existing knowledge editing methods for MLLMs lack the ability to handle multi-granularity knowledge and generalize to multimodal data. To address these limitations, we propose the multimodal knowledge editing method MLE. This approach identifies key knowledge layers within different components and collaboratively edits the various components of MLLMs. As a result, we observe significant improvements in visual generality performance, ranging from 4.8% to 10.8%, and achieve the best overall performance on knowledge data of different granularities.
UR - https://www.scopus.com/pages/publications/105040167041
U2 - 10.18653/v1/2025.emnlp-main.1478
DO - 10.18653/v1/2025.emnlp-main.1478
M3 - 会议稿件
AN - SCOPUS:105040167041
T3 - EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
SP - 29029
EP - 29042
BT - EMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference
A2 - Christodoulopoulos, Christos
A2 - Chakraborty, Tanmoy
A2 - Rose, Carolyn
A2 - Peng, Violet
PB - Association for Computational Linguistics (ACL)
Y2 - 4 November 2025 through 9 November 2025
ER -