Instruction Following on SelfInst (GPT-4 Feedback Score & Rouge-L)
21.7Rouge-L
Adversarial Moment-Matching Distillation
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| 21.7 | 54.2 | - | ||
2024.06 | 21.6 | 56.7 | - | |
2024.06 | 20.9 | 52.4 | - | |
2024.06 | 20.8 | 53.4 | - | |
2026.02 | 20.8 | - | 25.1 | |
2024.06 | 20.2 | 51.8 | - | |
2024.06 | 18.4 | 45 | - | |
2026.02 | 18.2 | - | 19 | |
2026.02 | 17.5 | - | 17.2 | |
2024.06 | 17.4 | 43.5 | - | |
2026.02 | 17.4 | - | 16.9 | |
2023.06 | 17.3 | 57.4 | - | |
2023.06 | 17.1 | 52.5 | - | |
2026.02 | 17.1 | - | 15.7 | |
2026.02 | 17 | - | 16.8 | |
2026.02 | 16.9 | - | 16.9 | |
2023.06 | 16.6 | 48.5 | - | |
2023.06 | 16.3 | 43.7 | - | |
2024.06 | 16.3 | 40.8 | - | |
2026.02 | 16.3 | - | 16 | |
2023.06 | 16.2 | 47.2 | - | |
2026.02 | 16.2 | - | 14.8 | |
2023.06 | 16 | 47 | - | |
2023.06 | 15.8 | 45.4 | - | |
2024.06 | 15.8 | 46.8 | - | |
2026.02 | 14.6 | - | 10.3 | |
2023.06 | 14.5 | 46 | - | |
2023.06 | 14.3 | 42.9 | - | |
2024.06 | 14.3 | 34.7 | - | |
2026.02 | 14.3 | - | 12.5 | |
2024.06 | 14.2 | 22.7 | - | |
2023.06 | 14 | 39.2 | - | |
2023.06 | 13.6 | 43.2 | - | |
2023.06 | 13.4 | 38.9 | - | |
2026.02 | 13.3 | - | 9.4 | |
2024.06 | 13.2 | 20.5 | - | |
2026.02 | 12.9 | - | 10 | |
2024.06 | 12.7 | 20.7 | - | |
2026.02 | 12.7 | - | 9.1 | |
2024.06 | 12.5 | 21.7 | - | |
2023.06 | 12.4 | 38.3 | - | |
2026.02 | 12.4 | - | 7 | |
2024.06 | 12.3 | 18 | - | |
2026.02 | 12.3 | - | 9.2 | |
2026.02 | 12.1 | - | 8.5 | |
2026.02 | 12.1 | - | 8.9 | |
2026.02 | 12 | - | 9.3 | |
2024.06 | 11.6 | 18.2 | - | |
2024.06 | 11.5 | 18.2 | - | |
2026.02 | 10.6 | - | 5.9 | |
2026.02 | 10.4 | - | 6.1 | |
2026.02 | 10.4 | - | 6.2 | |
2026.02 | 10.4 | - | 6.4 | |
2024.06 | 10.3 | 20.2 | - | |
2026.02 | 10.3 | - | 5.7 | |
2026.02 | 10.1 | - | 4.8 | |
2026.02 | 10 | - | 4.9 |