| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| VQAv2, GQA, VizWiz, SQA, TextVQA, POPE, MM-Bench, MM-Bench-CN | VQAv2 Accuracy80.7 | 9 | 1mo ago | ||
| Knowledge Datasets (ARC-C, ARC-E, MMLU, GPQA) | Reasoning | ARC-C64.75 | 9 | 4mo ago | |
| Knowledge Preservation Benchmark | ESD | FID33.3 | 8 | 4mo ago |