| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| G2U Utility | Mean Utility1 | 30 | 19d ago | ||
| MMLU | GPT-OSS-20B | Accuracy83.1 | 23 | 6d ago | |
| OPI clean | Gemma-4-E4B-it | Utility Score90.3 | 15 | 6d ago | |
| POPE | Neural Gate | Score85.56 | 14 | 2mo ago | |
| MME | DINM | Score0.7291 | 14 | 2mo ago | |
| ScienceQA | DINM | Score61.33 | 14 | 2mo ago |