| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Rep, Cod, and Req (test) | AutoDoS | Average Output Length16,384 | 18 | 5mo ago | |
| ToolSafe (test) | beam-search optimization framework | Think Length61,981 | 8 | 1mo ago | |
| MBPP | AiTM | ASR87.8 | 8 | 5mo ago | |
| HumanEval | AiTM | ASR63.8 | 8 | 5mo ago |