| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Scientific Code Generation | SciCode (test) | Subset Accuracy43.8 | 36 | |
| Scientific Coding | SciCode | Accuracy0.4275 | 18 | |
| Scientific Reasoning | SciCode | Accuracy43 | 10 | |
| Coding Agent Capacity | Scicode | Score57.7 | 8 | |
| Science Discovery | Scicode | Score57.7 | 8 | |
| Ranking Preservation | SciCode (test) | Mean Spearman Rho0.753 | 4 | |
| Code Reasoning | SciCode | Pass@142.1 | 4 | |
| Reasoning | SciCode | Score35.2 | 3 | |
| Reasoning | SciCode subtask | Score (SciCode Reasoning)39.6 | 1 |