Logical Reasoning on K&K Logistics (test)
0.69PPL2Qwen3-0.6B
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Qwen3-0.6BContext Length=2K, Training=-2026.02 | 0.69 | 0.47 | 0.41 | 0.22 | 0.09 | 0.08 | 0.03 | 0.28 | |
| Qwen3-0.6BContext Length=8K, Training=-2026.02 | 0.77 | 0.45 | 0.52 | 0.47 | 0.32 | 0.34 | 0.22 | 0.44 | |
| GRPOModel=Qwen3-0.6B, Context Length=2K2026.02 | 0.84 | 0.83 | 0.85 | 0.76 | 0.56 | 0.58 | 0.5 | 0.7 | |
| GRPOModel=Qwen3-0.6B, Context Length=8K2026.02 | 0.86 | 0.85 | 0.85 | 0.76 | 0.72 | 0.66 | 0.63 | 0.76 | |
| Qwen3-1.7BContext Length=2K, Training=-2026.02 | 0.9 | 0.7 | 0.66 | 0.43 | 0.21 | 0.16 | 0.05 | 0.44 | |
| Deepseek-R1Context Length=8K2026.02 | 0.91 | 0.73 | 0.77 | 0.78 | 0.75 | 0.88 | 0.83 | 0.81 | |
| Qwen3-1.7BContext Length=8K, Training=-2026.02 | 0.95 | 0.99 | 0.9 | 0.8 | 0.77 | 0.85 | 0.7 | 0.85 | |
| RLADModel=Qwen3-1.7B, Context Length=2K2026.02 | 0.98 | 0.99 | 1 | 0.98 | 0.93 | 0.86 | 0.75 | 0.93 | |
| o3-mini-highContext Length=8K2026.02 | 0.99 | 0.98 | 0.97 | 0.95 | 0.94 | 0.89 | 0.83 | 0.94 | |
| KDRLModel=Qwen3-1.7B, Context Length=8K2026.02 | 0.99 | 1 | 1 | 0.96 | 0.97 | 0.93 | 0.88 | 0.96 | |
| KDRLModel=Qwen3-0.6B, Context Length=8K2026.02 | 1 | 0.97 | 1 | 0.99 | 0.96 | 0.82 | 0.75 | 0.92 | |
| RLADModel=Qwen3-0.6B, Context Length=8K2026.02 | 1 | 0.99 | 0.99 | 0.96 | 0.93 | 0.91 | 0.83 | 0.94 | |
| GRPOModel=Qwen3-1.7B, Context Length=8K2026.02 | 1 | 0.99 | 1 | 0.98 | 0.95 | 0.92 | 0.81 | 0.95 | |
| RLADModel=Qwen3-1.7B, Context Length=8K2026.02 | 1 | 1 | 1 | 1 | 0.97 | 0.95 | 0.98 | 0.99 | |
| KDRLModel=Qwen3-0.6B, Context Length=2K2026.02 | 1 | 0.97 | 1 | 0.95 | 0.82 | 0.72 | 0.54 | 0.86 | |
| RLADModel=Qwen3-0.6B, Context Length=2K2026.02 | 1 | 1 | 0.99 | 0.96 | 0.86 | 0.81 | 0.69 | 0.9 | |
| GRPOModel=Qwen3-1.7B, Context Length=2K2026.02 | 1 | 0.98 | 0.97 | 0.9 | 0.87 | 0.79 | 0.52 | 0.86 | |
| KDRLModel=Qwen3-1.7B, Context Length=2K2026.02 | 1 | 0.98 | 1 | 0.98 | 0.86 | 0.84 | 0.69 | 0.91 |