Calibration
DeepSWE → NL2Repo
NL2Repo is estimated from DeepSWE with a linear curve fitted on 12 models measured on both: y = 0.3835·x + 0.3289, R² = 0.56, cross-validated error 6.6 pp. It is used for 3 estimates.
| Estimated model | DeepSWE | NL2Repo | Source |
|---|---|---|---|
| Pareto 26.10 Preview | 69.9% | 59.7% | estimated ± 6.6 pp, medium confidence |
| Pareto 26.9 | 74.0% | 61.3% | estimated ± 6.6 pp, medium confidence |
| SWE-2 | 73.0% | 60.9% | estimated ± 6.6 pp, medium confidence |