Code Reproduction on CPA Reproduction Stage
88.9Task Success RateAblateCell
Evaluation Results
| Method | Links | |
|---|---|---|
| AblateCellStage=Reproduction2026.04 | 88.9 | |
| Human ExpertsStage=Reproduction2026.04 | 50 | |
| RepoMasterStage=Reproduction2026.04 | 33.3 | |
| Mini-SWE-AgentStage=Reproduction2026.04 | 22.2 |