Visual Reasoning on CRIT Scientific Paper
15.9Exact Match (EM)Qwen2.5-VL_CRIT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen2.5-VL_CRIT# Prm=7B2026.04 | 15.9 | 22.5 | |
| Qwen3-VL-Thinking# Prm=32B2026.04 | 8.4 | 10.5 | |
| Kimi-VL-Thinking# Prm=16B2026.04 | 8.1 | 12.3 | |
| Intern3-VL# Prm=8B2026.04 | 7.3 | 12.1 | |
| Qwen3-VL-Thinking# Prm=8B2026.04 | 7.3 | 9.6 | |
| GLM-4.1V-Thinking# Prm=9B2026.04 | 5.2 | 10.4 |