Multimodal Mathematical Reasoning on MathVision (3040)
54.21Macro-average ScoreQwen3VL-8B-instruct + GeoSym Hard
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen3VL-8B-instruct + GeoSym HardBase Model=Qwen3-VL, Parameter Scale=8B, Method Component=GeoSym Hard2026.05 | 54.21 | 63.58 | 62.4 | 69.49 | |
| Qwen3VL-8B-instruct + GeoSym EntryBase Model=Qwen3-VL, Parameter Scale=8B, Method Component=GeoSym Entry2026.05 | 53.47 | 61.27 | 61.8 | 64.81 |