Software Issue Resolution on SWE-bench Lite
31.4Overall Resolution SuccessCADMAS-CTX
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CADMAS-CTX2026.04 | 31.4 | 33.5 | 24.8 | |
| Bucket-mean2026.04 | 26.5 | 32.8 | 18.2 | |
| Thompson Samp.2026.04 | 25.1 | 31.9 | 15.6 | |
| Static2026.04 | 22.3 | 32.1 | 8.4 |
| Method | Links | |||
|---|---|---|---|---|
| CADMAS-CTX2026.04 | 31.4 | 33.5 | 24.8 | |
| Bucket-mean2026.04 | 26.5 | 32.8 | 18.2 | |
| Thompson Samp.2026.04 | 25.1 | 31.9 | 15.6 | |
| Static2026.04 | 22.3 | 32.1 | 8.4 |