Rebuilding ROME : Resolving Model Collapse during Sequential Model Editing
Akshat Gupta, Sidharth Baskaran, Gopala Anumanchipalli
Abstract
Recent work using Rank-One Model Editing (ROME), a popular model editing method, has shown that there are certain facts that the algorithm is unable to edit without breaking the model. Such edits have previously been called disabling edits. These disabling edits cause immediate model collapse and limits the use of ROME for sequential editing. In this paper, we show that disabling edits are an artifact of irregularities in the implementation of ROME. With this paper, we provide a more stable implementation ROME, which we call r-ROME and show that model collapse is no longer observed when making large scale sequential edits with r-ROME, while further improving generalization and locality of model editing compared to the original implementation of ROME. We also provide a detailed mathematical explanation of the reason behind disabling edits.
BibTeX
@inproceedings{gupta-etal-2024-rebuilding,
title = "Rebuilding {ROME} : Resolving Model Collapse during Sequential Model Editing",
author = "Gupta, Akshat and
Baskaran, Sidharth and
Anumanchipalli, Gopala",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.emnlp-main.1210/",
doi = "10.18653/v1/2024.emnlp-main.1210",
pages = "21738--21744"
}