Multiple-Choice Classification on 14 standard NLP tasks suite (held-out)
86.9StoryCloze
OPT 175B
Updated 4d ago
Evaluation Results
| Method | Links | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2022.12 | 86.9 | 81.6 | 76.8 | 50.5 | 64.5 | 87.8 | 77.6 | 76.8 | 70 | 91.4 | 71 | 54.3 | 75.1 | 14 | 69.9 | |
2022.12 | 86.4 | 80.5 | 77.2 | 53.2 | 65 | 87.5 | 74.4 | 81.7 | 53.6 | 89 | 83.8 | 56.1 | 75.7 | 20.4 | 70.3 | |
2022.12 | 84.1 | 78.8 | 72.7 | 45.2 | 60.1 | 83.3 | 71.7 | 69.6 | 5.7 | 88.6 | 61.7 | 54 | 63.2 | 7.8 | 60.5 | |
2022.12 | 83.3 | 79.8 | 70.8 | 50.9 | 58.2 | 85.7 | 73 | 71.4 | 69.6 | 88 | 84.8 | 56.1 | 73 | 10.3 | 68.2 | |
2022.12 | 82.9 | 79.5 | 67 | 44.1 | 58.4 | 85.3 | 73.7 | 60.1 | 46.4 | 87 | 60.3 | 56.6 | 51.4 | 7.5 | 61.4 | |
2022.12 | 82.7 | 69.2 | 72.1 | 46.7 | 55.2 | 83.5 | 69 | 71.8 | 78.6 | 89 | 73.3 | 52 | 54.1 | 4.9 | 64.4 | |
2022.12 | 80.3 | 77.5 | 63.9 | 43.1 | 57.2 | 83.5 | 69.7 | 64 | 28.6 | 84 | 58.1 | 50.2 | 62.2 | 6.1 | 59.2 | |
2022.12 | 80.1 | 77.3 | 64.9 | 45.5 | 50.6 | 83.5 | 67.8 | 66.9 | 82.1 | 85 | 83.8 | 57.1 | 75.7 | 7.7 | 66.3 |