Factual Knowledge Question Answering on MMLU
83.3Accuracy
DeepSeek-V3
Updated 11d ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2026.04 | 83.3 | 56.3 | 5.2 | 27.5 | 93.2 | 99.5 | |
2026.04 | 77.1 | 59 | 16.6 | 35.3 | 85.3 | 98.7 | |
2026.04 | 71.6 | 53.9 | 7.3 | 19.8 | 89.9 | 98.8 | |
2026.04 | 71.2 | 63.9 | 42.6 | 56.8 | 63.5 | 92.5 | |
2026.04 | 62.8 | 64.8 | 32 | 83.5 | 49.4 | 94.1 | |
2026.04 | 62.4 | 62.9 | 32.9 | 65.7 | 58.3 | 94.9 | |
2026.04 | 56.3 | 65.4 | 47.9 | 84.4 | 40.5 | 89 |