TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/MatCha

MatCha

Reported on 10 benchmarks across 3 tasks · 1 paper · 8 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing8 results

  • Visual Question Answering (VQA)onPlotQA-D2
    1:1 Accuracy· 2022-12-19
    90.7
    best: 91.84 (MatCha4096 + LaMenDa)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Visual Question Answering (VQA)onPlotQA-D1
    1:1 Accuracy· 2022-12-19
    92.3
    best: 93.94 (MatCha4096 + LaMenDa)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Visual Question Answering (VQA)onPlotQA
    1:1 Accuracy· 2022-12-19
    91.5
    best: 92.89 (MatCha4096 + LaMenDa)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Visual Question Answering (VQA)onChartQA
    1:1 Accuracy· 2022-12-19
    64.2
    best: 81.3 (ChartPaLI-5B + PaLM 2-S)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Visual Question AnsweringonPlotQA-D2
    1:1 Accuracy· 2022-12-19
    90.7
    best: 91.84 (MatCha4096 + LaMenDa)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Visual Question AnsweringonPlotQA-D1
    1:1 Accuracy· 2022-12-19
    92.3
    best: 93.94 (MatCha4096 + LaMenDa)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Visual Question Answering (VQA)onDocVQA test
    ANLS· 2022-12-19
    0.742
    best: 0.9436 (Human)
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Visual Question Answering (VQA)onInfographicVQA
    ANLS· 2022-12-19
    37.2
    best: 80.3 (Gemini Ultra (pixel only))
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662

Computer Code2 results

  • Chart Question AnsweringonPlotQA
    1:1 Accuracy· 2022-12-19
    91.5
    best: 92.89 (MatCha4096 + LaMenDa)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662
  • Chart Question AnsweringonChartQA
    1:1 Accuracy· 2022-12-19
    64.2
    best: 81.3 (ChartPaLI-5B + PaLM 2-S)
    SOTA
    MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart DerenderingarXiv:2212.09662