INTERACTION: A Generative XAI Framework for Natural Language Inference Explanations

TL;DR

This paper introduces INTERACTION, a generative XAI framework for natural language inference that produces diverse, human-readable explanations and improves prediction accuracy, addressing the limitations of single-explanation approaches.

Contribution

The paper proposes a novel two-step generative framework that generates multiple diverse explanations and enhances prediction performance in NLP tasks.

Findings

Achieves up to 4.7% gain in BLEU for explanation quality

Achieves up to 4.4% gain in accuracy for predictions

Can generate multiple diverse explanations

Abstract

XAI with natural language processing aims to produce human-readable explanations as evidence for AI decision-making, which addresses explainability and transparency. However, from an HCI perspective, the current approaches only focus on delivering a single explanation, which fails to account for the diversity of human thoughts and experiences in language. This paper thus addresses this gap, by proposing a generative XAI framework, INTERACTION (explaIn aNd predicT thEn queRy with contextuAl CondiTional varIational autO-eNcoder). Our novel framework presents explanation in two steps: (step one) Explanation and Label Prediction; and (step two) Diverse Evidence Generation. We conduct intensive experiments with the Transformer architecture on a benchmark dataset, e-SNLI. Our method achieves competitive or better performance against state-of-the-art baseline models on explanation generation (up to 4.7% gain in BLEU) and prediction (up to 4.4% gain in accuracy) in step one; it can also generate multiple diverse explanations in step two.