Geometry Problem Solving on PyGeoX-Wild
81AccuracyProprietary-C
Evaluation Results
| Method | Links | |
|---|---|---|
| Proprietary-CInference=reasoning, Evaluation Protocol=zero-shot2026.06 | 81 | |
| Proprietary-BInference=reasoning, Evaluation Protocol=zero-shot2026.06 | 77 | |
| Qwen3-8B-RLInference=RL-trained, Evaluation Protocol=zero-shot2026.06 | 66 | |
| Proprietary-AInference=default, Evaluation Protocol=zero-shot2026.06 | 65 | |
| DeepSeek-V3.2Inference=default, Evaluation Protocol=zero-shot2026.06 | 64 |