Reasoning on AIME 2024 (Speedup, Mean acceptance length τ)
7.35SpeedupDFlash+DDTree
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DFlash+DDTreeModel=Qwen3-8B, Temperature=0.0, Node Budget=Best from {16, 32, 64, 128, 256, 512, 1024}2026.04 | 7.35 | 10.42 | |
| DFlash+DDTreeModel=Qwen3-4B, Temperature=0.0, Node Budget=Best from {16, 32, 64, 128, 256, 512, 1024}2026.04 | 7.27 | 10.37 | |
| DFlash+DDTreeModel=Qwen3-Coder-30B-A3B-Instruct, Temperature=0.0, Node Budget=Best from {16, 32, 64, 128, 256, 512, 1024}2026.04 | 5.93 | 7.87 | |
| DFlashModel=Qwen3-4B, Temperature=0.02026.04 | 5.56 | 7.54 | |
| DFlashModel=Qwen3-8B, Temperature=0.02026.04 | 5.38 | 7.46 | |
| DFlash+DDTreeModel=Qwen3-8B, Temperature=1.0, Node Budget=Best from {16, 32, 64, 128, 256, 512, 1024}2026.04 | 5.36 | 7.79 | |
| DFlash+DDTreeModel=Qwen3-4B, Temperature=1.0, Node Budget=Best from {16, 32, 64, 128, 256, 512, 1024}2026.04 | 5.31 | 7.82 | |
| DFlash+DDTreeModel=Qwen3-Coder-30B-A3B-Instruct, Temperature=1.0, Node Budget=Best from {16, 32, 64, 128, 256, 512, 1024}2026.04 | 4.94 | 7.32 | |
| DFlashModel=Qwen3-Coder-30B-A3B-Instruct, Temperature=0.02026.04 | 3.95 | 5.16 | |
| DFlashModel=Qwen3-4B, Temperature=1.02026.04 | 3.5 | 5.01 | |
| DFlashModel=Qwen3-8B, Temperature=1.02026.04 | 3.46 | 4.98 | |
| DFlashModel=Qwen3-Coder-30B-A3B-Instruct, Temperature=1.02026.04 | 3.12 | 4.23 |