| dataset,prompt,metric,value | |
| xnli_ar,GPT-3 style_arht,accuracy,0.41405622489959837 | |
| xnli_ar,MNLI crowdsource_arht,accuracy,0.3791164658634538 | |
| xnli_ar,can we infer_arht,accuracy,0.3530120481927711 | |
| xnli_ar,guaranteed/possible/impossible_arht,accuracy,0.42289156626506025 | |
| xnli_ar,justified in saying_arht,accuracy,0.3863453815261044 | |
| xnli_ar,median,accuracy,0.3863453815261044 | |
| xnli_es,GPT-3 style_esht,accuracy,0.44417670682730925 | |
| xnli_es,MNLI crowdsource_esht,accuracy,0.3337349397590361 | |
| xnli_es,can we infer_esht,accuracy,0.3333333333333333 | |
| xnli_es,guaranteed/possible/impossible_esht,accuracy,0.5682730923694779 | |
| xnli_es,justified in saying_esht,accuracy,0.3333333333333333 | |
| xnli_es,median,accuracy,0.3337349397590361 | |
| xnli_fr,GPT-3 style_frht,accuracy,0.44457831325301206 | |
| xnli_fr,MNLI crowdsource_frht,accuracy,0.3453815261044177 | |
| xnli_fr,can we infer_frht,accuracy,0.5847389558232932 | |
| xnli_fr,guaranteed/possible/impossible_frht,accuracy,0.4951807228915663 | |
| xnli_fr,justified in saying_frht,accuracy,0.5775100401606426 | |
| xnli_fr,median,accuracy,0.4951807228915663 | |
| xnli_hi,GPT-3 style_hiht,accuracy,0.3481927710843373 | |
| xnli_hi,MNLI crowdsource_hiht,accuracy,0.491566265060241 | |
| xnli_hi,can we infer_hiht,accuracy,0.44538152610441767 | |
| xnli_hi,guaranteed/possible/impossible_hiht,accuracy,0.6020080321285141 | |
| xnli_hi,justified in saying_hiht,accuracy,0.41124497991967873 | |
| xnli_hi,median,accuracy,0.44538152610441767 | |
| xnli_sw,GPT-3 style_swht,accuracy,0.3409638554216867 | |
| xnli_sw,MNLI crowdsource_swht,accuracy,0.3333333333333333 | |
| xnli_sw,can we infer_swht,accuracy,0.37710843373493974 | |
| xnli_sw,guaranteed/possible/impossible_swht,accuracy,0.3646586345381526 | |
| xnli_sw,justified in saying_swht,accuracy,0.3497991967871486 | |
| xnli_sw,median,accuracy,0.3497991967871486 | |
| xnli_ur,GPT-3 style_urht,accuracy,0.41767068273092367 | |
| xnli_ur,MNLI crowdsource_urht,accuracy,0.38433734939759034 | |
| xnli_ur,can we infer_urht,accuracy,0.3389558232931727 | |
| xnli_ur,guaranteed/possible/impossible_urht,accuracy,0.342570281124498 | |
| xnli_ur,justified in saying_urht,accuracy,0.3461847389558233 | |
| xnli_ur,median,accuracy,0.3461847389558233 | |
| xnli_vi,GPT-3 style_viht,accuracy,0.45461847389558235 | |
| xnli_vi,MNLI crowdsource_viht,accuracy,0.37269076305220883 | |
| xnli_vi,can we infer_viht,accuracy,0.39598393574297186 | |
| xnli_vi,guaranteed/possible/impossible_viht,accuracy,0.46987951807228917 | |
| xnli_vi,justified in saying_viht,accuracy,0.37028112449799194 | |
| xnli_vi,median,accuracy,0.39598393574297186 | |
| xnli_zh,GPT-3 style_zhht,accuracy,0.37710843373493974 | |
| xnli_zh,MNLI crowdsource_zhht,accuracy,0.3257028112449799 | |
| xnli_zh,can we infer_zhht,accuracy,0.3369477911646586 | |
| xnli_zh,guaranteed/possible/impossible_zhht,accuracy,0.3333333333333333 | |
| xnli_zh,justified in saying_zhht,accuracy,0.41365461847389556 | |
| xnli_zh,median,accuracy,0.3369477911646586 | |
| multiple,average,multiple,0.3861947791164659 | |