Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312055991> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4312055991 abstract "Visual language data such as plots, charts, and infographics are ubiquitous in the human world. However, state-of-the-art vision-language 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 image-to-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 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." @default.
- W4312055991 created "2023-01-04" @default.
- W4312055991 creator A5000738730 @default.
- W4312055991 creator A5003456660 @default.
- W4312055991 creator A5026154387 @default.
- W4312055991 creator A5043871919 @default.
- W4312055991 creator A5052308140 @default.
- W4312055991 creator A5073413742 @default.
- W4312055991 creator A5081690204 @default.
- W4312055991 creator A5081862885 @default.
- W4312055991 creator A5085471281 @default.
- W4312055991 date "2022-12-19" @default.
- W4312055991 modified "2023-09-26" @default.
- W4312055991 title "MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart Derendering" @default.
- W4312055991 doi "https://doi.org/10.48550/arxiv.2212.09662" @default.
- W4312055991 hasPublicationYear "2022" @default.
- W4312055991 type Work @default.
- W4312055991 citedByCount "0" @default.
- W4312055991 crossrefType "posted-content" @default.
- W4312055991 hasAuthorship W4312055991A5000738730 @default.
- W4312055991 hasAuthorship W4312055991A5003456660 @default.
- W4312055991 hasAuthorship W4312055991A5026154387 @default.
- W4312055991 hasAuthorship W4312055991A5043871919 @default.
- W4312055991 hasAuthorship W4312055991A5052308140 @default.
- W4312055991 hasAuthorship W4312055991A5073413742 @default.
- W4312055991 hasAuthorship W4312055991A5081690204 @default.
- W4312055991 hasAuthorship W4312055991A5081862885 @default.
- W4312055991 hasAuthorship W4312055991A5085471281 @default.
- W4312055991 hasBestOaLocation W43120559911 @default.
- W4312055991 hasConcept C105795698 @default.
- W4312055991 hasConcept C124101348 @default.
- W4312055991 hasConcept C127413603 @default.
- W4312055991 hasConcept C137293760 @default.
- W4312055991 hasConcept C138885662 @default.
- W4312055991 hasConcept C146978453 @default.
- W4312055991 hasConcept C154945302 @default.
- W4312055991 hasConcept C156365220 @default.
- W4312055991 hasConcept C190812933 @default.
- W4312055991 hasConcept C204321447 @default.
- W4312055991 hasConcept C2777055276 @default.
- W4312055991 hasConcept C2777508537 @default.
- W4312055991 hasConcept C2780878386 @default.
- W4312055991 hasConcept C33923547 @default.
- W4312055991 hasConcept C41008148 @default.
- W4312055991 hasConcept C41895202 @default.
- W4312055991 hasConceptScore W4312055991C105795698 @default.
- W4312055991 hasConceptScore W4312055991C124101348 @default.
- W4312055991 hasConceptScore W4312055991C127413603 @default.
- W4312055991 hasConceptScore W4312055991C137293760 @default.
- W4312055991 hasConceptScore W4312055991C138885662 @default.
- W4312055991 hasConceptScore W4312055991C146978453 @default.
- W4312055991 hasConceptScore W4312055991C154945302 @default.
- W4312055991 hasConceptScore W4312055991C156365220 @default.
- W4312055991 hasConceptScore W4312055991C190812933 @default.
- W4312055991 hasConceptScore W4312055991C204321447 @default.
- W4312055991 hasConceptScore W4312055991C2777055276 @default.
- W4312055991 hasConceptScore W4312055991C2777508537 @default.
- W4312055991 hasConceptScore W4312055991C2780878386 @default.
- W4312055991 hasConceptScore W4312055991C33923547 @default.
- W4312055991 hasConceptScore W4312055991C41008148 @default.
- W4312055991 hasConceptScore W4312055991C41895202 @default.
- W4312055991 hasLocation W43120559911 @default.
- W4312055991 hasOpenAccess W4312055991 @default.
- W4312055991 hasPrimaryLocation W43120559911 @default.
- W4312055991 hasRelatedWork W142374489 @default.
- W4312055991 hasRelatedWork W1803932089 @default.
- W4312055991 hasRelatedWork W1985007624 @default.
- W4312055991 hasRelatedWork W2176369193 @default.
- W4312055991 hasRelatedWork W2351428524 @default.
- W4312055991 hasRelatedWork W2359001871 @default.
- W4312055991 hasRelatedWork W3107474891 @default.
- W4312055991 hasRelatedWork W4312055991 @default.
- W4312055991 hasRelatedWork W970670907 @default.
- W4312055991 hasRelatedWork W2584532118 @default.
- W4312055991 isParatext "false" @default.
- W4312055991 isRetracted "false" @default.
- W4312055991 workType "article" @default.