Physics Reasoning on Synthetic Symbolic
10.4AccuracyQwen2.5-32B + RL (synthetic)
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-32B + RL (synthetic)Model=Qwen2.5-32B, Post-training Stage=RL (synthetic)2026.04 | 10.4 | |
| Qwen2.5-14B + RL (synthetic)Model=Qwen2.5-14B, Post-training Stage=RL (synthetic)2026.04 | 10.4 | |
| Qwen2.5-7B + RL (synthetic)Model=Qwen2.5-7B, Post-training Stage=RL (synthetic)2026.04 | 9.6 | |
| Qwen2.5-3B + RL (synthetic)Model=Qwen2.5-3B, Post-training Stage=RL (synthetic)2026.04 | 9.4 | |
| Qwen3-30BModel=Qwen3-30B, Post-training Stage=Instruct2026.04 | 8.8 | |
| Qwen3-30B + RL (synthetic)Model=Qwen3-30B, Post-training Stage=RL (synthetic)2026.04 | 8 | |
| Qwen2.5-32BModel=Qwen2.5-32B, Post-training Stage=Instruct2026.04 | 5.6 | |
| Qwen2.5-14BModel=Qwen2.5-14B, Post-training Stage=Instruct2026.04 | 5.6 | |
| Qwen2.5-7BModel=Qwen2.5-7B, Post-training Stage=Instruct2026.04 | 5.6 | |
| Qwen2.5-3BModel=Qwen2.5-3B, Post-training Stage=Instruct2026.04 | 3.2 |