| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Demo-ICL-Bench | Average Score80.1 | 14 | 3mo ago | ||
| DEMO | GPT-4o | Overall Score6.779 | 10 | 3mo ago | |
| Aggregate | openPangu-Embedded RL | Average Score68.73 | 9 | 26d ago | |
| Bilingual Full-Duplex-Bench English | SoulX-Duplug | Accuracy81.2 | 8 | 2mo ago | |
| BlenderBench | VIGA | Improvement159.19 | 8 | 3mo ago | |
| Bilingual Full-Duplex-Bench Chinese | SoulX-Duplug | Accuracy91.6 | 2 | 2mo ago |
