FAME: Towards Factual Multi-Task Model Editing

Li Zeng, Yingyu Shan, Zeming Liu, Jiashu Yao, Yuhang Guo


Abstract
Large language models (LLMs) embed extensive knowledge and utilize it to perform exceptionally well across various tasks. Nevertheless, outdated knowledge or factual errors within LLMs can lead to misleading or incorrect responses, causing significant issues in practical applications. To rectify the fatal flaw without the necessity for costly model retraining, various model editing approaches have been proposed to correct inaccurate information within LLMs in a cost-efficient way. To evaluate these model editing methods, previous work introduced a series of datasets. However, most of the previous datasets only contain fabricated data in a single format, which diverges from real-world model editing scenarios, raising doubts about their usability in practice. To facilitate the application of model editing in real-world scenarios, we propose the challenge of practicality. To resolve such challenges and effectively enhance the capabilities of LLMs, we present FAME, an authentic, comprehensive, and multi-task dataset, which is designed to enhance the practicality of model editing. We then propose SKEME, a model editing method that uses a novel caching mechanism to ensure synchronization with the real world. The experiments demonstrate that our method performs excellently across various tasks and scenarios, confirming its practicality.
Anthology ID:
2024.emnlp-main.894
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
15992–16011
Language:
URL:
https://aclanthology.org/2024.emnlp-main.894
DOI:
Bibkey:
Cite (ACL):
Li Zeng, Yingyu Shan, Zeming Liu, Jiashu Yao, and Yuhang Guo. 2024. FAME: Towards Factual Multi-Task Model Editing. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 15992–16011, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
FAME: Towards Factual Multi-Task Model Editing (Zeng et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.894.pdf