A 91-instance, human-screened Java SWE-bench set from six popular repositories, scored by fail-to-pass unit tests.
unassessed
| Category | coding |
|---|---|
| Subcategory | Java GitHub issue resolution / patch generation |
| Page status | active |
| Metric | resolved rate |
| Direction | higher_is_better |
| Unit | % |
| Dataset size | 91 |
| Dataset licence | Apache-2.0 |
| Publisher | Huawei and collaborators |
SWE-bench-java tests whether a model can resolve a real Java GitHub issue by patching the repository at the pre-fix commit. It is the Java analogue of SWE-bench Verified: issues are filtered so they compile, have fail-to-pass tests, and survive developer screening.
The system sees the issue text and a Dockerized Maven/Java checkout, then emits a patch. Tests from the fixing pull request grade the result. The authors evaluated SWE-agent rather than a one-shot RAG baseline.
No model card in ModelSpec reports this benchmark yet.