Detecting Edit Failures In Large Language Models: An Improved Specificity Benchmark
Jason Hoelscher-Obermaier, Julia Persson, Esben Kran, Ioannis Konstas, Fazl Barez
Abstract
Recent model editing techniques promise to mitigate the problem of memorizing false or outdated associations during LLM training. However, we show that these techniques can introduce large unwanted side effects which are not detected by existing specificity benchmarks. We extend the existing CounterFact benchmark to include a dynamic component and dub our benchmark CounterFact+. Additionally, we extend the metrics used for measuring specificity by a principled KL divergence-based metric. We use this improved benchmark to evaluate recent model editing techniques and find that they suffer from low specificity. Our findings highlight the need for improved specificity benchmarks that identify and prevent unwanted side effects.
BibTeX
@inproceedings{hoelscher-obermaier-etal-2023-detecting,
title = "Detecting Edit Failures In Large Language Models: An Improved Specificity Benchmark",
author = "Hoelscher-Obermaier, Jason and
Persson, Julia and
Kran, Esben and
Konstas, Ioannis and
Barez, Fazl",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.733/",
doi = "10.18653/v1/2023.findings-acl.733",
pages = "11548--11559"
}