| dataset,prompt,metric,value | |
| xnli_ar,GPT-3 style_arht,accuracy,0.4004016064257028 | |
| xnli_ar,MNLI crowdsource_arht,accuracy,0.41566265060240964 | |
| xnli_ar,can we infer_arht,accuracy,0.5670682730923695 | |
| xnli_ar,guaranteed/possible/impossible_arht,accuracy,0.5449799196787148 | |
| xnli_ar,justified in saying_arht,accuracy,0.5032128514056224 | |
| xnli_ar,median,accuracy,0.5032128514056224 | |
| xnli_es,GPT-3 style_esht,accuracy,0.44176706827309237 | |
| xnli_es,MNLI crowdsource_esht,accuracy,0.3333333333333333 | |
| xnli_es,can we infer_esht,accuracy,0.3333333333333333 | |
| xnli_es,guaranteed/possible/impossible_esht,accuracy,0.5875502008032129 | |
| xnli_es,justified in saying_esht,accuracy,0.3337349397590361 | |
| xnli_es,median,accuracy,0.3337349397590361 | |
| xnli_fr,GPT-3 style_frht,accuracy,0.4859437751004016 | |
| xnli_fr,MNLI crowdsource_frht,accuracy,0.38112449799196785 | |
| xnli_fr,can we infer_frht,accuracy,0.5550200803212851 | |
| xnli_fr,guaranteed/possible/impossible_frht,accuracy,0.39357429718875503 | |
| xnli_fr,justified in saying_frht,accuracy,0.5493975903614458 | |
| xnli_fr,median,accuracy,0.4859437751004016 | |
| xnli_hi,GPT-3 style_hiht,accuracy,0.4714859437751004 | |
| xnli_hi,MNLI crowdsource_hiht,accuracy,0.5220883534136547 | |
| xnli_hi,can we infer_hiht,accuracy,0.4461847389558233 | |
| xnli_hi,guaranteed/possible/impossible_hiht,accuracy,0.5172690763052209 | |
| xnli_hi,justified in saying_hiht,accuracy,0.42690763052208835 | |
| xnli_hi,median,accuracy,0.4714859437751004 | |
| xnli_sw,GPT-3 style_swht,accuracy,0.3437751004016064 | |
| xnli_sw,MNLI crowdsource_swht,accuracy,0.3337349397590361 | |
| xnli_sw,can we infer_swht,accuracy,0.3502008032128514 | |
| xnli_sw,guaranteed/possible/impossible_swht,accuracy,0.3453815261044177 | |
| xnli_sw,justified in saying_swht,accuracy,0.3477911646586345 | |
| xnli_sw,median,accuracy,0.3453815261044177 | |
| xnli_ur,GPT-3 style_urht,accuracy,0.38313253012048193 | |
| xnli_ur,MNLI crowdsource_urht,accuracy,0.3891566265060241 | |
| xnli_ur,can we infer_urht,accuracy,0.46626506024096387 | |
| xnli_ur,guaranteed/possible/impossible_urht,accuracy,0.3614457831325301 | |
| xnli_ur,justified in saying_urht,accuracy,0.46947791164658637 | |
| xnli_ur,median,accuracy,0.3891566265060241 | |
| xnli_vi,GPT-3 style_viht,accuracy,0.3353413654618474 | |
| xnli_vi,MNLI crowdsource_viht,accuracy,0.43373493975903615 | |
| xnli_vi,can we infer_viht,accuracy,0.3337349397590361 | |
| xnli_vi,guaranteed/possible/impossible_viht,accuracy,0.48032128514056227 | |
| xnli_vi,justified in saying_viht,accuracy,0.336144578313253 | |
| xnli_vi,median,accuracy,0.336144578313253 | |
| xnli_zh,GPT-3 style_zhht,accuracy,0.5369477911646586 | |
| xnli_zh,MNLI crowdsource_zhht,accuracy,0.3473895582329317 | |
| xnli_zh,can we infer_zhht,accuracy,0.5325301204819277 | |
| xnli_zh,guaranteed/possible/impossible_zhht,accuracy,0.334136546184739 | |
| xnli_zh,justified in saying_zhht,accuracy,0.5289156626506024 | |
| xnli_zh,median,accuracy,0.5289156626506024 | |
| multiple,average,multiple,0.4242469879518072 | |