| Benchmark | Model | Rank | Results |
|---|---|---|---|
| arithmetic-reasoning-on-gsm8k | Qwen3-8B-Base with Reasoning Palette (RLOO, Linear Decay) | – | Accuracy: 93.03 |
| mathematical-reasoning-on-math-500 | Qwen3-8B-Base with Reasoning Palette (RLOO, Linear Decay) | – | Accuracy: 72.2 |
| mathematical-reasoning-on-minerva-math | Qwen3-8B-Base with Reasoning Palette (RLOO, Linear Decay) | – | Accuracy: 43.77 |
| mathematical-reasoning-on-olympiadbench | Qwen3-8B-Base with Reasoning Palette (RLOO, Linear Decay) | – | Accuracy: 46.61 |