QADiver: Interactive Framework for Diagnosing QA Models

TL;DR

QADiver is an interactive web-based framework designed to diagnose and understand the performance of question answering models by visualizing their contributions and explanations, aiding researchers in model refinement.

Contribution

This paper introduces QADiver, a novel interactive framework that visualizes and analyzes QA model behavior to facilitate diagnosis and improvement.

Findings

Enables detailed visualization of model contributions

Helps identify sources of QA errors

Supports model refinement through interactive analysis

Abstract

Question answering (QA) extracting answers from text to the given question in natural language, has been actively studied and existing models have shown a promise of outperforming human performance when trained and evaluated with SQuAD dataset. However, such performance may not be replicated in the actual setting, for which we need to diagnose the cause, which is non-trivial due to the complexity of model. We thus propose a web-based UI that provides how each model contributes to QA performances, by integrating visualization and analysis tools for model explanation. We expect this framework can help QA model researchers to refine and improve their models.