Mathematical Reasoning on GSM8K (Accuracy, TKN)
94.8AccuracyCLSR
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CLSRBackbone=Qwen3-8B, T (number of rounds)=32026.06 | 94.8 | 214 | |
| PBrdBackbone=Qwen3-8B2026.06 | 93.6 | 235 | |
| CLSRBackbone=Qwen3-8B2026.06 | 92.8 | 90 | |
| PALBackbone=Qwen3-8B2026.06 | 92.5 | 148 | |
| P2SBackbone=Qwen3-8B2026.06 | 92.2 | 316 | |
| PoTBackbone=Qwen3-8B2026.06 | 92.1 | 113 | |
| CLSRBackbone=Qwen3-8B, T (number of rounds)=12026.06 | 92.1 | 83 | |
| CoTBackbone=Qwen3-8B2026.06 | 91.6 | 248 |