On Explaining Your Explanations Of BERT: An Empirical Study With Sequence Classification

Wu Zhengxuan, Ong Desmond C.. Arxiv 2021

[Paper]    
Attention Mechanism BERT Fine Tuning Interpretability And Explainability Model Architecture Pretraining Methods Security Training Techniques

BERT, as one of the pretrianed language models, attracts the most attention in recent years for creating new benchmarks across GLUE tasks via fine-tuning. One pressing issue is to open up the blackbox and explain the decision makings of BERT. A number of attribution techniques have been proposed to explain BERT models, but are often limited to sequence to sequence tasks. In this paper, we adapt existing attribution methods on explaining decision makings of BERT in sequence classification tasks. We conduct extensive analyses of four existing attribution methods by applying them to four different datasets in sentiment analysis. We compare the reliability and robustness of each method via various ablation studies. Furthermore, we test whether attribution methods explain generalized semantics across semantically similar tasks. Our work provides solid guidance for using attribution methods to explain decision makings of BERT for downstream classification tasks.

Similar Work