SlideVQA: A Dataset for Document Visual Question Answering on Multiple Images
This dataset addresses the problem of multi-image reasoning in document VQA for researchers, though it is incremental as it extends existing single-image datasets.
The authors introduced SlideVQA, a dataset for document visual question answering across multiple slide images, containing 2.6k+ slide decks, 52k+ images, and 14.5k questions requiring complex reasoning. Their new model outperformed existing state-of-the-art QA models but still lags behind human performance.
Visual question answering on document images that contain textual, visual, and layout information, called document VQA, has received much attention recently. Although many datasets have been proposed for developing document VQA systems, most of the existing datasets focus on understanding the content relationships within a single image and not across multiple images. In this study, we propose a new multi-image document VQA dataset, SlideVQA, containing 2.6k+ slide decks composed of 52k+ slide images and 14.5k questions about a slide deck. SlideVQA requires complex reasoning, including single-hop, multi-hop, and numerical reasoning, and also provides annotated arithmetic expressions of numerical answers for enhancing the ability of numerical reasoning. Moreover, we developed a new end-to-end document VQA model that treats evidence selection and question answering in a unified sequence-to-sequence format. Experiments on SlideVQA show that our model outperformed existing state-of-the-art QA models, but that it still has a large gap behind human performance. We believe that our dataset will facilitate research on document VQA.