| dataset,prompt,metric,value | |
| xnli_ar,GPT-3 style_arht,accuracy,0.40441767068273093 | |
| xnli_ar,MNLI crowdsource_arht,accuracy,0.43012048192771085 | |
| xnli_ar,can we infer_arht,accuracy,0.3610441767068273 | |
| xnli_ar,guaranteed/possible/impossible_arht,accuracy,0.3642570281124498 | |
| xnli_ar,justified in saying_arht,accuracy,0.37309236947791163 | |
| xnli_ar,median,accuracy,0.37309236947791163 | |
| xnli_es,GPT-3 style_esht,accuracy,0.5698795180722892 | |
| xnli_es,MNLI crowdsource_esht,accuracy,0.342570281124498 | |
| xnli_es,can we infer_esht,accuracy,0.46546184738955826 | |
| xnli_es,guaranteed/possible/impossible_esht,accuracy,0.5526104417670683 | |
| xnli_es,justified in saying_esht,accuracy,0.4321285140562249 | |
| xnli_es,median,accuracy,0.46546184738955826 | |
| xnli_fr,GPT-3 style_frht,accuracy,0.4995983935742972 | |
| xnli_fr,MNLI crowdsource_frht,accuracy,0.4004016064257028 | |
| xnli_fr,can we infer_frht,accuracy,0.5694779116465863 | |
| xnli_fr,guaranteed/possible/impossible_frht,accuracy,0.5152610441767068 | |
| xnli_fr,justified in saying_frht,accuracy,0.5493975903614458 | |
| xnli_fr,median,accuracy,0.5152610441767068 | |
| xnli_hi,GPT-3 style_hiht,accuracy,0.44417670682730925 | |
| xnli_hi,MNLI crowdsource_hiht,accuracy,0.5236947791164659 | |
| xnli_hi,can we infer_hiht,accuracy,0.4963855421686747 | |
| xnli_hi,guaranteed/possible/impossible_hiht,accuracy,0.4493975903614458 | |
| xnli_hi,justified in saying_hiht,accuracy,0.4963855421686747 | |
| xnli_hi,median,accuracy,0.4963855421686747 | |
| xnli_sw,GPT-3 style_swht,accuracy,0.39397590361445783 | |
| xnli_sw,MNLI crowdsource_swht,accuracy,0.3329317269076305 | |
| xnli_sw,can we infer_swht,accuracy,0.4285140562248996 | |
| xnli_sw,guaranteed/possible/impossible_swht,accuracy,0.38433734939759034 | |
| xnli_sw,justified in saying_swht,accuracy,0.41967871485943775 | |
| xnli_sw,median,accuracy,0.39397590361445783 | |
| xnli_ur,GPT-3 style_urht,accuracy,0.463855421686747 | |
| xnli_ur,MNLI crowdsource_urht,accuracy,0.40441767068273093 | |
| xnli_ur,can we infer_urht,accuracy,0.3895582329317269 | |
| xnli_ur,guaranteed/possible/impossible_urht,accuracy,0.3405622489959839 | |
| xnli_ur,justified in saying_urht,accuracy,0.43293172690763054 | |
| xnli_ur,median,accuracy,0.40441767068273093 | |
| xnli_vi,GPT-3 style_viht,accuracy,0.5261044176706827 | |
| xnli_vi,MNLI crowdsource_viht,accuracy,0.39879518072289155 | |
| xnli_vi,can we infer_viht,accuracy,0.5481927710843374 | |
| xnli_vi,guaranteed/possible/impossible_viht,accuracy,0.43694779116465865 | |
| xnli_vi,justified in saying_viht,accuracy,0.46546184738955826 | |
| xnli_vi,median,accuracy,0.46546184738955826 | |
| xnli_zh,GPT-3 style_zhht,accuracy,0.36947791164658633 | |
| xnli_zh,MNLI crowdsource_zhht,accuracy,0.3457831325301205 | |
| xnli_zh,can we infer_zhht,accuracy,0.3441767068273092 | |
| xnli_zh,guaranteed/possible/impossible_zhht,accuracy,0.4923694779116466 | |
| xnli_zh,justified in saying_zhht,accuracy,0.3927710843373494 | |
| xnli_zh,median,accuracy,0.36947791164658633 | |
| multiple,average,multiple,0.4354417670682731 | |