{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.02469", "kind": "paper", "url": "https://arxiv.org/abs/2607.02469"}, {"kind": "website", "url": "https://www.testevo-bench.com"}], "authors": ["Jiale Amber Wang", "Kaiyuan Wang", "Pengyu Nie"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.02469", "source_url": "https://arxiv.org/abs/2607.02469"}, "categories": ["Agents", "Code & Software", "General AI"], "description": {"en": "TestEvo-Bench evaluates test and code co-evolution tasks from real commit histories. It includes two tracks: test generation (write new tests for changed behavior) and test update (adapt failing tests). Tasks are packaged with environment configurations for execution-grounded metrics like pass rate, coverage, and mutation score. The live component uses timestamps to enable post-training-cutoff evaluation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.02469", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.02469", "title": "TestEvo-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of Waterloo", "Google (United States)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.02469", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.02469", "source_url": "https://openalex.org/W7167250826"}, "key": "claire-radar:2607.02469", "modality": null, "name": "TestEvo-Bench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_testevo-bench_296298e8", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "064e97b0901826614c288053248a7ebaf6bdbd8e9ec408cd760426f78240ab30", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.02469"}, "publisher": null, "released": "2026-07-02", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.02469", "source_url": "https://arxiv.org/abs/2607.02469"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-02469", "source": "claire_radar", "source_benchmark_id": "2607.02469", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Code & Software", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2607.02469", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "f56c6e3473f4e16d55190c712fc8c8ec2a11b35ad6ca6e4170c335464af40a5d", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "TestEvo-Bench evaluates test and code co-evolution tasks from real commit histories. It includes two tracks: test generation (write new tests for changed behavior) and test update (adapt failing tests). Tasks are packaged with environment configurations for execution-grounded metrics like pass rate, coverage, and mutation score. The live component uses timestamps to enable post-training-cutoff evaluation.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce TestEvo-Bench, a benchmark of test and code co-evolution tasks mined from software repositories, with two tracks: in test generation, the agent shall write new tests to capture the new software behavior; in test update, the agent shall adapt failing existing tests to the changed software behavior."}, "familyId": "bmf_b73f50c8cb68", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_testevo-bench_296298e8", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2607.02469", "paper": "https://arxiv.org/abs/2607.02469", "pdf": "https://arxiv.org/pdf/2607.02469", "project": "https://www.testevo-bench.com", "report": "https://arxiv.org/abs/2607.02469"}, "metrics": [], "motivation": "Software tests and code evolve together: a code change should be followed by new or updated tests that record the new software behavior.", "name": "TestEvo-Bench", "oneLine": "TestEvo-Bench evaluates test and code co-evolution tasks from real commit histories. It includes two tracks: test generation (write new tests for changed behavior) and test update (adapt failing tests). Tasks are packaged with environment configurations for execution-grounded metrics like pass rate, coverage, and mutation score. The live component uses timestamps to enable post-training-cutoff evaluation.", "paperTitle": "TestEvo-Bench: An Executable and Live Benchmark for Test and Code Co-Evolution", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.02469", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-02", "source": {"authors": ["Jiale Amber Wang", "Kaiyuan Wang", "Pengyu Nie"], "categories": ["cs.SE", "cs.AI", "cs.CL"], "id": "2607.02469", "title": "TestEvo-Bench: An Executable and Live Benchmark for Test and Code Co-Evolution", "type": "arxiv", "url": "https://arxiv.org/abs/2607.02469"}, "sourceUpdatedAt": "2026-07-02T17:35:20Z", "topics": ["Agents"], "venueAttempts": [], "whyItMatters": "Existing benchmarks often separate tests from code changes, relying on static metadata without verifying executability. TestEvo-Bench provides a dynamic, execution-grounded evaluation that reflects real-world test maintenance, enabling assessment of test automation agents in tracking code changes. Its live design reduces data leakage risk, making it valuable for comparing agents in practical settings."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
