Tell Model Where to Attend: Improving Interpretability of Aspect-Based Sentiment Classification via Small Explanation Annotations

TL;DR

This paper introduces IEGA, a gradient-based interpretability framework for aspect-based sentiment classification that uses minimal explanation annotations to improve model focus, interpretability, performance, and robustness.

Contribution

The paper proposes a novel gradient correction framework, IEGA, which enhances interpretability and performance of ABSC models with small annotation effort.

Findings

Improves interpretability of ABSC models.

Enhances model robustness and accuracy.

Applicable to various models and tasks.

Abstract

Gradient-based explanation methods play an important role in the field of interpreting complex deep neural networks for NLP models. However, the existing work has shown that the gradients of a model are unstable and easily manipulable, which impacts the model's reliability largely. According to our preliminary analyses, we also find the interpretability of gradient-based methods is limited for complex tasks, such as aspect-based sentiment classification (ABSC). In this paper, we propose an \textbf{I}nterpretation-\textbf{E}nhanced \textbf{G}radient-based framework for \textbf{A}BSC via a small number of explanation annotations, namely \texttt{{IEGA}}. Particularly, we first calculate the word-level saliency map based on gradients to measure the importance of the words in the sentence towards the given aspect. Then, we design a gradient correction module to enhance the model's attention on the correct parts (e.g., opinion words). Our model is model agnostic and task agnostic so that it can be integrated into the existing ABSC methods or other tasks. Comprehensive experimental results on four benchmark datasets show that our \texttt{IEGA} can improve not only the interpretability of the model but also the performance and robustness.