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/MDETR

MDETR

Reported on 11 benchmarks across 2 tasks · 1 paper · 3 SOTA

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

Natural Language Processing9 results

  • Visual Question Answering (VQA)onCLEVR-Humans
    Accuracy· 2021-04-26
    81.7
    SOTA
    MDETR -- Modulated Detection for End-to-End Multi-Modal UnderstandingarXiv:2104.12763
  • Visual Question Answering (VQA)onCLEVR
    Accuracy· 2021-04-26
    99.7
    best: 99.8 (NS-VQA (1K programs))
    MDETR -- Modulated Detection for End-to-End Multi-Modal UnderstandingarXiv:2104.12763
  • Visual Question Answering (VQA)onGQA Test2019
    Accuracy
    62.45
    best: 89.3 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Binary
    80.91
    best: 91.2 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Consistency
    93.95
    best: 98.4 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Distribution
    5.36
    best: 93.08 (GlobalPrior)
  • Visual Question Answering (VQA)onGQA Test2019
    Open
    46.15
    best: 87.4 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Plausibility
    84.15
    best: 97.2 (human)
  • Visual Question Answering (VQA)onGQA Test2019
    Validity
    96.33
    best: 98.9 (human)

Computer Vision2 results

  • Generalized Referring Expression ComprehensionongRefCOCO
    N-acc.· 2021-04-26
    36.1
    best: 54.7 (SimVG-DB)
    SOTA
    MDETR -- Modulated Detection for End-to-End Multi-Modal UnderstandingarXiv:2104.12763
  • Generalized Referring Expression ComprehensionongRefCOCO
    Precision@(F1=1, IoU≥0.5)· 2021-04-26
    41.5
    best: 62.1 (SimVG-DB)
    SOTA
    MDETR -- Modulated Detection for End-to-End Multi-Modal UnderstandingarXiv:2104.12763