Multimodal Scientific Claim Verification on SciClaimEval Table (dev)
81.7F1 ScoreQwen3-VL-30B-A3B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-VL-30B-A3BFramework=SciLens Agent2026.06 | 81.7 | 67.4 | |
| Qwen3-VL-30B-A3BFramework=Vanilla LLM (baseline)2026.06 | 80.6 | 62.7 | |
| Qwen3-VL-8BFramework=SciLens Agent2026.06 | 80.5 | 61.9 | |
| Qwen3-VL-8BFramework=Vanilla LLM (baseline)2026.06 | 75.4 | 51.7 | |
| Qwen3-VL-4BFramework=Vanilla LLM (baseline)2026.06 | 74.2 | 49.2 | |
| InternVL3_5-14BFramework=Vanilla LLM (baseline)2026.06 | 72 | 49.2 | |
| InternVL3_5-38BFramework=Vanilla LLM (baseline)2026.06 | 71.7 | 47.5 | |
| InternVL3_5-8BFramework=Vanilla LLM (baseline)2026.06 | 70.6 | 44.5 |