| Benchmark | Model | Rank | Results |
|---|---|---|---|
| common-sense-reasoning-on-winogrande | KiC-770M | – | Accuracy: 55.30 |
| coreference-resolution-on-winograd-schema | KiC-770M | – | Accuracy: 65.40 |
| natural-language-inference-on-anli-test | KiC-770M | – | A1: 36.30A2: 35.00A3: 37.60 |
| natural-language-inference-on-rte | KiC-770M | – | Accuracy: 74.00 |
| question-answering-on-copa | KiC-770M | – | Accuracy: 85.30 |
| question-answering-on-storycloze | KiC-770M | – | Accuracy: 94.40 |
| sentence-completion-on-hellaswag | KiC-770M | – | Accuracy: 29.6 |
| word-sense-disambiguation-on-words-in-context | KiC-770M | – | Accuracy: 52.40 |