Multimodal Scientific Claim Verification on SciClaimEval Figure (dev)
74.3F1 ScoreQwen3-VL-30B-A3B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-VL-30B-A3BFramework=SciLens Agent2026.06 | 74.3 | 54.3 | |
| Qwen3-VL-8BFramework=SciLens Agent2026.06 | 67.7 | 42.2 | |
| Qwen3-VL-30B-A3BFramework=Vanilla LLM (baseline)2026.06 | 66.6 | 38.8 | |
| Qwen3-VL-8BFramework=Vanilla LLM (baseline)2026.06 | 65.8 | 37.1 | |
| Qwen3-VL-4BFramework=Vanilla LLM (baseline)2026.06 | 63.7 | 32.8 | |
| InternVL3_5-14BFramework=Vanilla LLM (baseline)2026.06 | 59.7 | 25.9 | |
| InternVL3_5-38BFramework=Vanilla LLM (baseline)2026.06 | 57.9 | 25 | |
| InternVL3_5-8BFramework=Vanilla LLM (baseline)2026.06 | 52.2 | 16.4 |