DocVXQA: Context-Aware Visual Explanations for Document Question Answering

Mohamed Ali Souibgui, Changkyu Choi, Andrey Barsky, Kangsoo Jung, Ernest Valveny, Dimosthenis Karatzas

Abstract:

We propose DocVXQA, a novel framework for visually self-explainable document question answering. The framework is designed not only to produce accurate answers to questions but also to learn visual heatmaps that highlight contextually critical regions, thereby offering interpretable justifications for the model’s decisions. To integrate explanations into the learning process, we quantitatively formulate explainability principles as explicit learning objectives. Unlike conventional methods that emphasize only the regions pertinent to the answer, our framework delivers explanations that are contextually sufficient while remaining representation-efficient. This fosters user
trust while achieving a balance between predictive performance and interpretability in DocVQA applications. Extensive experiments, including human evaluation, provide strong evidence supporting the effectiveness of our method. The code is available at https://github.com/dali92002/DocVXQA.