| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Unweighted Average | II | ECE9.8 | 29 | 21d ago | |
| LLaVA Evaluation Suite (MMBench, MME, MM-Vet, ScienceQA) 1.5 (test val) | LoRA | MMBench68.5 | 16 | 3mo ago | |
| DRBench BS | SCI7 | MCQ Score29.68 | 14 | 2mo ago | |
| DRBench S Subset | M3ID | MCQ Accuracy47.22 | 14 | 2mo ago | |
| DRBench B | SCI7 | MCQ Score27.04 | 14 | 2mo ago | |
| SEED | CogVLM-Chat | Overall Score72.5 | 13 | 3mo ago | |
| MME | HTDC | Perception Score1,711.44 | 12 | 1mo ago | |
| MMBench v1.0 (dev) | CogVLM-Chat | Score77.6 | 11 | 3mo ago | |
| LLAVA (bench) | CogVLM-Chat | Score77.8 | 10 | 3mo ago |