Flexible Model Interpretability through Natural Language Model Editing

2023/11/17 by D'Oosterlinck, Karel, Demeester, Thomas, Develder, Chris +1
#Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2311.10905

Abstract

Model interpretability and model editing are crucial goals in the age of large language models. Interestingly, there exists a link between these two goals: if a method is able to systematically edit model behavior with regard to a human concept of interest, this editor method can help make internal representations more interpretable by pointing towards relevant representations and systematically manipulating them.

Related

• Enhancing Semantic Consistency of Large Language Models through Model Editing: An Interpretability-Oriented Approach
• Should We Really Edit Language Models? On the Evaluation of Edited Language Models
• Model Editing Harms General Abilities of Large Language Models: Regularization to the Rescue
• An Investigation of Language Model Interpretability via Sentence Editing
• Is Model Editing Built on Sand? Revealing Its Illusory Success and Fragile Foundation