{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.04167", "kind": "paper", "url": "https://arxiv.org/abs/2609.04167"}, {"id": "gh:deepsoftwareanalytics/swe-gate", "kind": "repo", "url": "https://github.com/DeepSoftwareAnalytics/SWE-Gate"}], "authors": ["Xin He", "Yanlin Wang", "Mingwei Liu", "Jiachi Chen", "Hongyu Zhang", "Guanbin Li"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.04167", "source_url": "https://arxiv.org/abs/2609.04167"}, "categories": ["Code", "Code & Software", "General AI", "Software & AI Compute"], "description": {"en": "Evaluates repository-level repair agents on both functional correctness and adherence to review-derived constraints, using 303 instances with separate functional and constraint tests."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.04167", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.04167", "title": "SWE-Gate"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.04167", "modality": null, "name": "SWE-Gate", "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": "true", "origin_record_id": "bm_swe-gate_9dd200ce", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "4acca6a6c05450a80ba5636bc5761b2a67af69940528841cc9fdb0c378136fd1", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-07T00:42:15.394147Z", "source_url": "https://arxiv.org/abs/2609.04167"}, "publisher": null, "released": "2026-09-03", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.04167", "source_url": "https://arxiv.org/abs/2609.04167"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-04167", "source": "claire_radar", "source_benchmark_id": "2609.04167", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI", "Software & AI Compute"], "area": "Code & Software", "attention": {"asOf": "2026-09-04", "attemptedAt": "2026-09-04T06:36:46.783164Z", "githubRepo": "https://github.com/DeepSoftwareAnalytics/SWE-Gate", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2609.04167", "observedAt": "2026-09-04T06:36:46.783164Z", "signalStatus": {"githubStars": {"observedAt": "2026-09-04T06:36:46.783164Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-09-04T06:36:46.783164Z", "state": "unavailable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses a widely recognized gap in software engineering agent evaluation and provides strong artifacts, making it likely to attract attention from the agent benchmarking and software engineering communities.", "score": 68}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-07T00:42:15.394147Z", "inputHash": "8cea50d2b3a2b6cfd05b3702488d88db1d63a7ef986674cf0634640eae7cfebf", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "We introduce SWE-Gate, a repository-level benchmark for software engineering agents that explicitly evaluates review constraint compliance alongside functional correctness.", "canonicalNameSource": "abstract", "decisionReason": "Includes a released canonical dataset, evaluation tools, and a clear scoring contract separating functional and constraint tests, with public replication materials.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-07T00:42:15.394147Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates repository-level repair agents on both functional correctness and adherence to review-derived constraints, using 303 instances with separate functional and constraint tests.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce SWE-Gate, a repository-level benchmark for software engineering agents that explicitly evaluates review constraint compliance alongside functional correctness."}, "familyId": "bmf_00e8694fdf58", "firstSeenAt": "2026-09-04", "heat": null, "id": "bm_swe-gate_9dd200ce", "indexedAt": "2026-09-04T06:13:54.443310Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/DeepSoftwareAnalytics/SWE-Gate", "data": null, "paper": "https://arxiv.org/abs/2609.04167", "pdf": "https://arxiv.org/pdf/2609.04167", "project": null, "report": "https://arxiv.org/abs/2609.04167"}, "metrics": [], "motivation": "Repository-level software engineering benchmarks have significantly advanced the evaluation of coding agents, but existing benchmarks primarily measure whether generated patches pass functional tests and overlook review-derived acceptance constraints (review constraints) that often influence whether a patch is acceptable in real-world software development.", "name": "SWE-Gate", "oneLine": "Evaluates repository-level repair agents on both functional correctness and adherence to review-derived constraints, using 303 instances with separate functional and constraint tests.", "paperTitle": "SWE-Gate: Passing Functional Tests Is Not Enough for Software Engineering Agents", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.04167", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-04T06:13:54.443310Z"}, "publications": [], "publishers": [{"name": "DeepSoftwareAnalytics", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/DeepSoftwareAnalytics/SWE-Gate"}], "ranking": {"30d": {"confidence": "Low", "coverage": 0.55, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.0, "value": 0}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 122, "score": 23}, "method": "attention-ranking-v13", "rank": 122, "score": 23}, "90d": {"confidence": "Low", "coverage": 0.55, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.0, "value": 0}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 318, "score": 23}, "method": "attention-ranking-v13", "rank": 318, "score": 23}, "today": {"confidence": "Low", "coverage": 0.25, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": 1, "historyDate": "2026-09-03", "rank": null, "score": null, "windowDays": 1}, "level": {"components": {"githubStars": {"normalization": "percentile", "normalized": 0.2857142857142857, "percentile": 0.2857142857142857, "value": 0}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "llmAttentionForecast": {"maxContribution": 0.25, "percentile": 0.72, "role": "bonus", "value": 72}}, "confidence": "Low", "coverage": 0.25, "rank": 7, "score": 50}, "method": "attention-ranking-v13", "rank": 7, "score": 50}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-09-03", "source": {"authors": ["Xin He", "Yanlin Wang", "Mingwei Liu", "Jiachi Chen", "Hongyu Zhang", "Guanbin Li"], "categories": ["cs.SE", "cs.AI"], "id": "2609.04167", "title": "SWE-Gate: Passing Functional Tests Is Not Enough for Software Engineering Agents", "type": "arxiv", "url": "https://arxiv.org/abs/2609.04167"}, "sourceUpdatedAt": "2026-09-03T17:53:34Z", "topics": ["Code"], "venueAttempts": [], "whyItMatters": "Reveals that functional tests alone overestimate agent readiness for real-world patch acceptance, highlighting the need to measure review constraint compliance in software engineering evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
