Natural Language Inference on ANLI Round 2 (test)
51.4AccuracyInfoBERT
Evaluation Results
| Method | Links | |
|---|---|---|
| InfoBERTlearning_mode=Supervised2022.01 | 51.4 | |
| GLaMModel Size=64B/64E, Evaluation Protocol=Few-shot, Shots=102021.12 | 41.2 | |
| GLaMModel Size=64B/64E, Evaluation Protocol=One-shot2021.12 | 40 | |
| MT-NLGshot_setting=One-shot2022.01 | 39.7 | |
| MT-NLGshot_setting=Few-shot2022.01 | 39.6 | |
| Megatron-NLGModel Size=530B, Evaluation Protocol=Few-shot2021.12 | 39.6 | |
| GLaMModel Size=64B/64E, Evaluation Protocol=Zero-shot2021.12 | 37.3 | |
| MT-NLGshot_setting=Zero-shot2022.01 | 36.6 | |
| GPT-3shot_setting=Zero-shot2022.01 | 35.4 | |
| GPT-3Model Size=175B, Evaluation Protocol=Zero-shot2021.12 | 35.4 | |
| GPT-3shot_setting=Few-shot2022.01 | 34 | |
| GPT-3Model Size=175B, Evaluation Protocol=Few-shot, Shots=502021.12 | 34 | |
| GPT-3shot_setting=One-shot2022.01 | 33.9 | |
| GPT-3Model Size=175B, Evaluation Protocol=One-shot2021.12 | 33.9 |