Interpreting Pretrained Language Models Via Concept Bottlenecks · The Large Language Model Bible Contribute to LLM-Bible

Interpreting Pretrained Language Models Via Concept Bottlenecks

Tan Zhen, Cheng Lu, Wang Song, Bo Yuan, Li Jundong, Liu Huan. Arxiv 2023

[Paper]    
Attention Mechanism Interpretability And Explainability Model Architecture Reinforcement Learning Security Survey Paper Transformer

Pretrained language models (PLMs) have made significant strides in various natural language processing tasks. However, the lack of interpretability due to their black-box'' nature poses challenges for responsible implementation. Although previous studies have attempted to improve interpretability by using, e.g., attention weights in self-attention layers, these weights often lack clarity, readability, and intuitiveness. In this research, we propose a novel approach to interpreting PLMs by employing high-level, meaningful concepts that are easily understandable for humans. For example, we learn the concept of Food’’ and investigate how it influences the prediction of a model’s sentiment towards a restaurant review. We introduce C\(^3\)M, which combines human-annotated and machine-generated concepts to extract hidden neurons designed to encapsulate semantically meaningful and task-specific concepts. Through empirical evaluations on real-world datasets, we manifest that our approach offers valuable insights to interpret PLM behavior, helps diagnose model failures, and enhances model robustness amidst noisy concept labels.

Similar Work