A Suite of Generative Tasks for Multi-Level Multimodal Webpage Understanding
This addresses the problem of underused structured image-text data for researchers in multimodal AI, though it is incremental as it builds on existing webpage resources.
The authors tackled the lack of comprehensive multimodal webpage datasets by introducing WikiWeb2M, a suite with 2M pages containing image, text, and structure data, and demonstrated its utility on generative tasks like page description generation, showing improved performance over prior work.
Webpages have been a rich, scalable resource for vision-language and language only tasks. Yet only pieces of webpages are kept in existing datasets: image-caption pairs, long text articles, or raw HTML, never all in one place. Webpage tasks have resultingly received little attention and structured image-text data left underused. To study multimodal webpage understanding, we introduce the Wikipedia Webpage suite (WikiWeb2M) containing 2M pages with all of the associated image, text, and structure data. We verify its utility on three generative tasks: page description generation, section summarization, and contextual image captioning. We design a novel attention mechanism Prefix Global, which selects the most relevant image and text content as global tokens to attend to the rest of the webpage for context. By using page structure to separate such tokens, it performs better than full attention with lower computational complexity. Extensive experiments show that the new data in WikiWeb2M improves task performance compared to prior work.