Text-rich Visual Comprehension on SEED-Bench 2-Plus
70.22AccuracyReGuLaR-7B
Evaluation Results
| Method | Links | |
|---|---|---|
| ReGuLaR-7BModel Category=Latent reasoning LVLMs2026.05 | 70.22 | |
| LaserModel Category=Latent reasoning LVLMs2026.05 | 70.05 | |
| DeepeyesModel Category=RL-based reasoning LVLMs2026.05 | 69.08 | |
| Vision-R1Model Category=RL-based reasoning LVLMs2026.05 | 68.95 | |
| MonetModel Category=Latent reasoning LVLMs2026.05 | 65.88 | |
| Qwen2.5-VL-7BModel Category=General-purpose LVLMs2026.05 | 65.31 | |
| LLaVA-OneVisionModel Category=General-purpose LVLMs2026.05 | 61.22 | |
| PAPOModel Category=RL-based reasoning LVLMs2026.05 | 54.11 | |
| LVRModel Category=Latent reasoning LVLMs2026.05 | 47.39 |