List your company

Question Answering on SQuADShifts 1.0 (test)

54.3Average F1multitask QAG
16.02825.96435.945.836May 26, 2023

Evaluation results

MethodLinks
multitask QAGLanguage Model=BART-Large2023.0554.336.753.634.354.136.457.741.851.634.4
end2end QAGLanguage Model=T5-Large2023.0553.737.349.131.156.140.160.745.548.832.5
Gold QA2023.0553.337.345.930.455.638.761.446.950.133.4
pipeline QAGLanguage Model=T5-Large2023.055233.950.931.151.732.955.939.549.732
end2end QAGLanguage Model=T5-Base2023.0551.835.444.926.956.940.159.945.345.329.2
pipeline QAGLanguage Model=BART-Large2023.0551.7344930.252.533.755.34049.732.3
pipeline QAGLanguage Model=T5-Base2023.0551.733.650.631.152.83453.837.149.632.4
multitask QAGLanguage Model=BART-Base2023.0550.833.249.430.550.632.15539.248.431.1
QG onlyLanguage Model=T5-Base2023.0550.734.743.327.454.437.157.743.247.331.1
pipeline QAGLanguage Model=BART-Base2023.055032.448.429.849.431.1533649.532.7
multitask QAGLanguage Model=T5-Base2023.0549.630.948.828.948.128.552.535.149.131.2
multitask QAGLanguage Model=T5-Large2023.0549.530.946.32649.630.65335.94931.2
QG onlyLanguage Model=BART-Base2023.0549.433.942.326.754.337.259.344.841.927
QG onlyLanguage Model=BART-Large2023.0549.433.843.327.45436.759.444.641.126.4
QG onlyLanguage Model=T5-Large2023.0548.933.442.726.853.236.258.543.941.526.7
QG onlyLanguage Model=T5-Small2023.0548.533.143.827.750.534.555.24144.429.1
end2end QAGLanguage Model=T5-Small2023.0548.331.742.324.854.737.255.240.141.124.8
pipeline QAGLanguage Model=T5-Small2023.0545.427.441.623.846.927.248.932.144.126.6
multitask QAGLanguage Model=T5-Small2023.054425.242.122.944.324.145.627.94425.7
end2end QAGLanguage Model=BART-Base2023.053421.429.316.535.423.244.631.126.615
end2end QAGLanguage Model=BART-Large2023.0517.510.217.19.118.911.322.314.711.65.7