MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart Derendering

Visual language data such as plots, charts, and infographics are ubiquitous in the human world. However, state-of-the-art visionlanguage models do not perform well on these data. We propose MATCHA (Math reasoning and Chart derendering pretraining) to enhance visual language models' capabilities in jointly modeling charts/plots and language data. Specifically we propose several pretraining tasks that cover plot deconstruction and numerical reasoning which are the key capabilities in visual language modeling. We perform the MATCHA pretraining starting from Pix2Struct, a recently proposed imageto-text visual language model. On standard benchmarks such as PlotQA and ChartQA, the MATCHA model outperforms state-of-the-art methods by as much as nearly 20%. We also examine how well the MATCHA pretraining transfers to domains such as screenshots, textbook diagrams, and document figures and observe overall improvement, verifying the usefulness of MATCHA pretraining on broader visual language tasks. 12