Logical Reasoning on Countdown CD34
78.2Avg@16DS
Evaluation Results
| Method | Links | |
|---|---|---|
| DSBackbone=Qwen3-8B, Finetuning=Yes, Runtime=6.9h2026.02 | 78.2 | |
| GPSBackbone=Qwen3-8B, Finetuning=Yes, Runtime=5.0h2026.02 | 77.9 | |
| DSBackbone=Qwen3-4B, Finetuning=Yes, Runtime=4.9h2026.02 | 76.3 | |
| GPSBackbone=Qwen3-4B, Finetuning=Yes, Runtime=3.4h2026.02 | 76 | |
| MoPPSBackbone=Qwen3-8B, Finetuning=Yes, Runtime=4.3h2026.02 | 76 | |
| GRESOBackbone=Qwen3-8B, Finetuning=Yes, Runtime=6.2h2026.02 | 75.1 | |
| PCLBackbone=Qwen3-8B, Finetuning=Yes, Runtime=5.1h2026.02 | 74.9 | |
| MoPPSBackbone=Qwen3-4B, Finetuning=Yes, Runtime=2.8h2026.02 | 73.9 | |
| UniformBackbone=Qwen3-4B, Finetuning=Yes, Runtime=2.9h2026.02 | 73.8 | |
| GRESOBackbone=Qwen3-4B, Finetuning=Yes, Runtime=4.9h2026.02 | 73.8 | |
| UniformBackbone=Qwen3-8B, Finetuning=Yes, Runtime=4.3h2026.02 | 73.3 | |
| PCLBackbone=Qwen3-4B, Finetuning=Yes, Runtime=3.2h2026.02 | 72.8 | |
| Qwen3-8BBackbone=Qwen3-8B, Finetuning=None2026.02 | 3.9 | |
| Qwen3-4BBackbone=Qwen3-4B, Finetuning=None2026.02 | 3.5 |