Explanatory Visual Question Answering

Explanatory Visual Question Answering (EVQA) is an advanced task in the field of computer vision that aims to answer visual questions and generate multimodal explanations to reveal the reasoning process. This task not only requires accurate understanding of image content but also demands the integration of natural language and visual elements to comprehensively express the logic of reasoning, thereby enhancing the transparency and interpretability of the model. It holds significant application value.