General Reasoning on ScienceQA
89.4AccuracyDUEL
Evaluation Results
| Method | Links | |
|---|---|---|
| DUELRole=Solver2026.05 | 89.4 | |
| VLAA-Thinker-7B2026.05 | 88.9 | |
| OpenVLThinker-7B2026.05 | 88.6 | |
| DUELRole=Challenger2026.05 | 88.5 | |
| EvoLMM2026.05 | 88.4 | |
| MM-UPT2026.05 | 88.3 | |
| Vision-Zero(CLEVR)2026.05 | 88.2 | |
| Qwen2.5-VL-7B2026.05 | 88.1 |