Multi-turn Dialogue on MT-Bench (Acceptance Length τ and Speedup Ratio)
5.78Acceptance Length (τ)
PPOW
Updated 19d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.05 | 5.78 | 3.54 | |
2026.05 | 5.58 | 3.02 | |
2026.05 | 5.53 | 2.91 | |
2026.05 | 5.47 | 2.72 | |
2026.05 | 5.45 | 3.73 | |
2026.05 | 5.25 | 3.47 | |
2026.05 | 5.14 | 2.58 | |
2026.05 | 5.14 | 3.52 | |
2026.05 | 5.12 | 3.63 | |
2026.05 | 5.08 | 3.6 | |
2026.05 | 4.95 | 2.64 | |
2026.05 | 4.95 | 2.81 | |
2026.05 | 4.82 | 3.4 | |
2026.05 | 4.72 | 2.82 | |
2026.05 | 4.7 | 3.38 | |
2026.05 | 4.51 | 2.6 | |
2026.05 | 4.17 | 2.35 | |
2026.05 | 4.12 | 2.29 | |
2026.05 | 4.03 | 2.24 | |
2026.05 | 4 | 2.41 |