Model

T5-11B

Vendor N/A

Knowledge & Reasoning

Knowledge & Reasoning
BenchmarkScoreMetricHarnessSourceConfidenceLink
BoolQ91.2scoreN/AIndependentmediumarxiv.org
CNN/Daily Mail43.5rouge-1N/AIndependentmediumarxiv.org
MRPC90.0accuracyN/AIndependentmediumarxiv.org
QNLI96.7accuracyN/AIndependentmediumarxiv.org
Quora Question Pairs90.4accuracyN/AIndependentmediumarxiv.org
SuperGLUE88.9scoreN/AIndependentmediumarxiv.org

Multimodal

Multimodal
BenchmarkScoreMetricHarnessSourceConfidenceLink
CoLA70.8accuracyN/AIndependentmediumarxiv.org

Reproduced by the community

Loading community runs…