Scientific QA on SCIMDR Eval
49.9AccuracyGPT-5.2
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-5.2Model=GPT-5.22026.03 | 49.9 | |
| Qwen2.5-VL-7B + SCIMDRModel=Qwen2.5-VL-7B, Training Data=SCIMDR2026.03 | 49.1 | |
| GPT-5.1Model=GPT-5.12026.03 | 47.2 | |
| Qwen-3-VL-8BModel=Qwen-3-VL-8B2026.03 | 34.2 | |
| Qwen2.5-VL-7BModel=Qwen2.5-VL-7B, Training Data=None2026.03 | 19.8 | |
| InternVL-3-8BModel=InternVL-3-8B2026.03 | 16.8 | |
| LLaVA-OV-1.5-8BModel=LLaVA-OV-1.5-8B2026.03 | 15.5 | |
| Qwen2.5-VL-7B + SPIQAModel=Qwen2.5-VL-7B, Training Data=SPIQA2026.03 | 5.6 |