dataset,prompt,metric,value
anli_dev_r1,GPT-3 style,accuracy,0.426
anli_dev_r1,MNLI crowdsource,accuracy,0.402
anli_dev_r1,can we infer,accuracy,0.401
anli_dev_r1,guaranteed/possible/impossible,accuracy,0.314
anli_dev_r1,justified in saying,accuracy,0.387
anli_dev_r1,median,accuracy,0.401
anli_dev_r2,GPT-3 style,accuracy,0.383
anli_dev_r2,MNLI crowdsource,accuracy,0.374
anli_dev_r2,can we infer,accuracy,0.394
anli_dev_r2,guaranteed/possible/impossible,accuracy,0.302
anli_dev_r2,justified in saying,accuracy,0.376
anli_dev_r2,median,accuracy,0.376
anli_dev_r3,GPT-3 style,accuracy,0.42
anli_dev_r3,MNLI crowdsource,accuracy,0.4116666666666667
anli_dev_r3,can we infer,accuracy,0.38916666666666666
anli_dev_r3,guaranteed/possible/impossible,accuracy,0.2966666666666667
anli_dev_r3,justified in saying,accuracy,0.35833333333333334
anli_dev_r3,median,accuracy,0.38916666666666666
story_cloze_2016,Answer Given options,accuracy,0.8524853019775521
story_cloze_2016,Choose Story Ending,accuracy,0.8957776590058792
story_cloze_2016,Generate Ending,accuracy,0.709246392303581
story_cloze_2016,Novel Correct Ending,accuracy,0.8888295029396045
story_cloze_2016,Story Continuation and Options,accuracy,0.8850881881346874
story_cloze_2016,median,accuracy,0.8850881881346874
super_glue_cb,GPT-3 style,accuracy,0.8392857142857143
super_glue_cb,MNLI crowdsource,accuracy,0.35714285714285715
super_glue_cb,can we infer,accuracy,0.7857142857142857
super_glue_cb,guaranteed/possible/impossible,accuracy,0.5535714285714286
super_glue_cb,justified in saying,accuracy,0.7142857142857143
super_glue_cb,median,accuracy,0.7142857142857143
super_glue_copa,"C1 or C2? premise, so/because…",accuracy,0.66
super_glue_copa,best_option,accuracy,0.77
super_glue_copa,cause_effect,accuracy,0.8
super_glue_copa,i_am_hesitating,accuracy,0.81
super_glue_copa,plausible_alternatives,accuracy,0.84
super_glue_copa,median,accuracy,0.8
super_glue_rte,GPT-3 style,accuracy,0.7906137184115524
super_glue_rte,MNLI crowdsource,accuracy,0.8267148014440433
super_glue_rte,does it follow that,accuracy,0.7942238267148014
super_glue_rte,guaranteed true,accuracy,0.776173285198556
super_glue_rte,should assume,accuracy,0.7617328519855595
super_glue_rte,median,accuracy,0.7906137184115524
winogrande_winogrande_xl,Replace,accuracy,0.5588003157063931
winogrande_winogrande_xl,True or False,accuracy,0.5280189423835833
winogrande_winogrande_xl,does underscore refer to,accuracy,0.5651144435674822
winogrande_winogrande_xl,stand for,accuracy,0.5082872928176796
winogrande_winogrande_xl,underscore refer to,accuracy,0.5651144435674822
winogrande_winogrande_xl,median,accuracy,0.5588003157063931
xcopa_id,"C1 or C2? premise, so/because…",accuracy,0.46
xcopa_id,best_option,accuracy,0.7
xcopa_id,cause_effect,accuracy,0.73
xcopa_id,i_am_hesitating,accuracy,0.72
xcopa_id,plausible_alternatives,accuracy,0.67
xcopa_id,median,accuracy,0.7
xcopa_sw,"C1 or C2? premise, so/because…",accuracy,0.6
xcopa_sw,best_option,accuracy,0.55
xcopa_sw,cause_effect,accuracy,0.54
xcopa_sw,i_am_hesitating,accuracy,0.51
xcopa_sw,plausible_alternatives,accuracy,0.52
xcopa_sw,median,accuracy,0.54
xcopa_ta,"C1 or C2? premise, so/because…",accuracy,0.59
xcopa_ta,best_option,accuracy,0.56
xcopa_ta,cause_effect,accuracy,0.6
xcopa_ta,i_am_hesitating,accuracy,0.57
xcopa_ta,plausible_alternatives,accuracy,0.62
xcopa_ta,median,accuracy,0.59
xcopa_vi,"C1 or C2? premise, so/because…",accuracy,0.53
xcopa_vi,best_option,accuracy,0.72
xcopa_vi,cause_effect,accuracy,0.72
xcopa_vi,i_am_hesitating,accuracy,0.7
xcopa_vi,plausible_alternatives,accuracy,0.71
xcopa_vi,median,accuracy,0.71
xcopa_zh,"C1 or C2? premise, so/because…",accuracy,0.67
xcopa_zh,best_option,accuracy,0.7
xcopa_zh,cause_effect,accuracy,0.8
xcopa_zh,i_am_hesitating,accuracy,0.77
xcopa_zh,plausible_alternatives,accuracy,0.79
xcopa_zh,median,accuracy,0.77
xnli_ar,GPT-3 style,accuracy,0.5558232931726907
xnli_ar,MNLI crowdsource,accuracy,0.42128514056224897
xnli_ar,can we infer,accuracy,0.5148594377510041
xnli_ar,guaranteed/possible/impossible,accuracy,0.40562248995983935
xnli_ar,justified in saying,accuracy,0.4927710843373494
xnli_ar,median,accuracy,0.4927710843373494
xnli_en,GPT-3 style,accuracy,0.5891566265060241
xnli_en,MNLI crowdsource,accuracy,0.42610441767068274
xnli_en,can we infer,accuracy,0.5662650602409639
xnli_en,guaranteed/possible/impossible,accuracy,0.4614457831325301
xnli_en,justified in saying,accuracy,0.5437751004016064
xnli_en,median,accuracy,0.5437751004016064
xnli_es,GPT-3 style,accuracy,0.5734939759036145
xnli_es,MNLI crowdsource,accuracy,0.40923694779116465
xnli_es,can we infer,accuracy,0.5148594377510041
xnli_es,guaranteed/possible/impossible,accuracy,0.43132530120481927
xnli_es,justified in saying,accuracy,0.4610441767068273
xnli_es,median,accuracy,0.4610441767068273
xnli_fr,GPT-3 style,accuracy,0.5666666666666667
xnli_fr,MNLI crowdsource,accuracy,0.42208835341365464
xnli_fr,can we infer,accuracy,0.5385542168674698
xnli_fr,guaranteed/possible/impossible,accuracy,0.39076305220883534
xnli_fr,justified in saying,accuracy,0.5100401606425703
xnli_fr,median,accuracy,0.5100401606425703
xnli_hi,GPT-3 style,accuracy,0.5345381526104418
xnli_hi,MNLI crowdsource,accuracy,0.41124497991967873
xnli_hi,can we infer,accuracy,0.4751004016064257
xnli_hi,guaranteed/possible/impossible,accuracy,0.40923694779116465
xnli_hi,justified in saying,accuracy,0.4469879518072289
xnli_hi,median,accuracy,0.4469879518072289
xnli_sw,GPT-3 style,accuracy,0.4827309236947791
xnli_sw,MNLI crowdsource,accuracy,0.40562248995983935
xnli_sw,can we infer,accuracy,0.44497991967871486
xnli_sw,guaranteed/possible/impossible,accuracy,0.42289156626506025
xnli_sw,justified in saying,accuracy,0.41124497991967873
xnli_sw,median,accuracy,0.42289156626506025
xnli_ur,GPT-3 style,accuracy,0.4947791164658635
xnli_ur,MNLI crowdsource,accuracy,0.39759036144578314
xnli_ur,can we infer,accuracy,0.4502008032128514
xnli_ur,guaranteed/possible/impossible,accuracy,0.39036144578313253
xnli_ur,justified in saying,accuracy,0.40843373493975904
xnli_ur,median,accuracy,0.40843373493975904
xnli_vi,GPT-3 style,accuracy,0.5449799196787148
xnli_vi,MNLI crowdsource,accuracy,0.40401606425702813
xnli_vi,can we infer,accuracy,0.5
xnli_vi,guaranteed/possible/impossible,accuracy,0.44779116465863456
xnli_vi,justified in saying,accuracy,0.4650602409638554
xnli_vi,median,accuracy,0.4650602409638554
xnli_zh,GPT-3 style,accuracy,0.5429718875502008
xnli_zh,MNLI crowdsource,accuracy,0.3891566265060241
xnli_zh,can we infer,accuracy,0.5032128514056224
xnli_zh,guaranteed/possible/impossible,accuracy,0.38072289156626504
xnli_zh,justified in saying,accuracy,0.4706827309236948
xnli_zh,median,accuracy,0.4706827309236948
xstory_cloze_ar,Answer Given options,accuracy,0.6896095301125083
xstory_cloze_ar,Choose Story Ending,accuracy,0.8378557246856386
xstory_cloze_ar,Generate Ending,accuracy,0.5956320317670417
xstory_cloze_ar,Novel Correct Ending,accuracy,0.8213103904698875
xstory_cloze_ar,Story Continuation and Options,accuracy,0.8219722038385175
xstory_cloze_ar,median,accuracy,0.8213103904698875
xstory_cloze_es,Answer Given options,accuracy,0.7683653209794837
xstory_cloze_es,Choose Story Ending,accuracy,0.886168100595632
xstory_cloze_es,Generate Ending,accuracy,0.6724023825281271
xstory_cloze_es,Novel Correct Ending,accuracy,0.8676373262739907
xstory_cloze_es,Story Continuation and Options,accuracy,0.8769027134348114
xstory_cloze_es,median,accuracy,0.8676373262739907
xstory_cloze_eu,Answer Given options,accuracy,0.6082064857710126
xstory_cloze_eu,Choose Story Ending,accuracy,0.7266710787557908
xstory_cloze_eu,Generate Ending,accuracy,0.5552614162806089
xstory_cloze_eu,Novel Correct Ending,accuracy,0.700198544010589
xstory_cloze_eu,Story Continuation and Options,accuracy,0.7107875579086698
xstory_cloze_eu,median,accuracy,0.700198544010589
xstory_cloze_hi,Answer Given options,accuracy,0.6366644606221046
xstory_cloze_hi,Choose Story Ending,accuracy,0.7882197220383852
xstory_cloze_hi,Generate Ending,accuracy,0.5982792852415619
xstory_cloze_hi,Novel Correct Ending,accuracy,0.7485109199205824
xstory_cloze_hi,Story Continuation and Options,accuracy,0.7683653209794837
xstory_cloze_hi,median,accuracy,0.7485109199205824
xstory_cloze_id,Answer Given options,accuracy,0.7385837193911317
xstory_cloze_id,Choose Story Ending,accuracy,0.8332230311052283
xstory_cloze_id,Generate Ending,accuracy,0.6293845135671741
xstory_cloze_id,Novel Correct Ending,accuracy,0.7816015883520847
xstory_cloze_id,Story Continuation and Options,accuracy,0.8226340172071476
xstory_cloze_id,median,accuracy,0.7816015883520847
xstory_cloze_zh,Answer Given options,accuracy,0.7498345466578424
xstory_cloze_zh,Choose Story Ending,accuracy,0.8583719391131701
xstory_cloze_zh,Generate Ending,accuracy,0.6227663798808736
xstory_cloze_zh,Novel Correct Ending,accuracy,0.8405029781601588
xstory_cloze_zh,Story Continuation and Options,accuracy,0.8385175380542687
xstory_cloze_zh,median,accuracy,0.8385175380542687
xwinograd_en,Replace,accuracy,0.6576344086021505
xwinograd_en,True or False,accuracy,0.5187096774193548
xwinograd_en,does underscore refer to,accuracy,0.5931182795698925
xwinograd_en,stand for,accuracy,0.5070967741935484
xwinograd_en,underscore refer to,accuracy,0.6210752688172043
xwinograd_en,median,accuracy,0.5931182795698925
xwinograd_fr,Replace,accuracy,0.5180722891566265
xwinograd_fr,True or False,accuracy,0.5301204819277109
xwinograd_fr,does underscore refer to,accuracy,0.5542168674698795
xwinograd_fr,stand for,accuracy,0.5180722891566265
xwinograd_fr,underscore refer to,accuracy,0.5421686746987951
xwinograd_fr,median,accuracy,0.5301204819277109
xwinograd_pt,Replace,accuracy,0.5741444866920152
xwinograd_pt,True or False,accuracy,0.4790874524714829
xwinograd_pt,does underscore refer to,accuracy,0.55893536121673
xwinograd_pt,stand for,accuracy,0.5209125475285171
xwinograd_pt,underscore refer to,accuracy,0.5437262357414449
xwinograd_pt,median,accuracy,0.5437262357414449
xwinograd_zh,Replace,accuracy,0.626984126984127
xwinograd_zh,True or False,accuracy,0.503968253968254
xwinograd_zh,does underscore refer to,accuracy,0.5436507936507936
xwinograd_zh,stand for,accuracy,0.49007936507936506
xwinograd_zh,underscore refer to,accuracy,0.5535714285714286
xwinograd_zh,median,accuracy,0.5436507936507936
multiple,average,multiple,0.6067197952551315