{"record": {"aliases": [], "artifacts": [{"id": "gh:json604/silentbug-bench", "kind": "repo", "url": "https://github.com/Json604/silentbug-bench"}], "categories": ["Language & Knowledge", "Software & AI Compute", "cs.AI"], "description": {"en": "Evaluates agents on detecting silent ML training defects across 18 CPU-only PyTorch tasks, scoring patch application, metric recovery, localization, minimality, and false positives."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://github.com/Json604/silentbug-bench", "source": "claire_radar", "source_url": "https://github.com/Json604/silentbug-bench", "title": "silentbug"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:github:json604/silentbug-bench", "modality": null, "name": "silentbug", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-candidate", "display_eligible": "false", "origin_record_id": "bm_silentbug-bench_29b06a59", "origin_source": "github", "recognition_confidence": "0.4", "record_sha256": "9228bbfb2fd0f3805ffa21996e849e00e013b2cfe2316b73cc524d75367859aa", "relation": "unclear", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-08-25T17:43:56.718736Z", "source_url": "https://github.com/Json604/silentbug-bench"}, "publisher": null, "released": "2026-08-20", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:github:json604/silentbug-bench", "source_url": "https://github.com/Json604/silentbug-bench"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-github-json604-silentbug-bench", "source": "claire_radar", "source_benchmark_id": "github:json604/silentbug-bench", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Software & AI Compute"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/Json604/silentbug-bench", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": null, "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"state": "not_applicable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses an underexplored problem in ML debugging with clear evaluation mechanics, likely to interest researchers in code agents and ML engineering.", "score": 50}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-25T17:43:56.718736Z", "inputHash": "25c69576ef23b0e8839b78ea572984cfee4ff1107dd27ed3504f7178a9a00493", "model": "deepseek-v4-pro"}, "curation": {"decisionReason": "The formal benchmark name is source-grounded, but the public release does not yet meet the independent evidence or adoption threshold.", "model": "deepseek-v4-pro", "reviewedAt": "2026-08-25T17:43:56.718736Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-candidate", "demo": false, "description": "Evaluates agents on detecting silent ML training defects across 18 CPU-only PyTorch tasks, scoring patch application, metric recovery, localization, minimality, and false positives.", "discoverySources": [{"id": "github:json604/silentbug-bench", "type": "github", "url": "https://github.com/Json604/silentbug-bench"}], "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["discovered via github", "benchmark term in abstract", "evaluation protocol evidence", "no explicit benchmark release evidence"], "snippet": "silentbug-bench ML training defects that never crash."}, "familyId": "bmf_9af0df30b045", "firstSeenAt": "2026-08-23", "heat": null, "id": "bm_silentbug-bench_29b06a59", "indexedAt": "2026-08-23T06:55:51.988821Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/Json604/silentbug-bench"}, "metrics": [], "motivation": "silentbug-bench ML training defects that never crash.", "name": "silentbug", "oneLine": "Evaluates agents on detecting silent ML training defects across 18 CPU-only PyTorch tasks, scoring patch application, metric recovery, localization, minimality, and false positives.", "paperTitle": "silentbug-bench", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://github.com/Json604/silentbug-bench", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-23T06:55:51.988821Z"}, "publications": [], "publishers": [{"name": "Json604", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Json604/silentbug-bench"}], "ranking": {"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": 516, "score": 23}, "method": "attention-ranking-v13", "rank": 516, "score": 23}}, "readiness": "Runnable", "recognitionConfidence": 0.4, "relation": "unclear", "releasedAt": "2026-08-20", "source": {"authors": ["Json604"], "categories": [], "id": "github:json604/silentbug-bench", "title": "silentbug-bench", "type": "github", "url": "https://github.com/Json604/silentbug-bench"}, "sourceUpdatedAt": "2026-08-20T01:02:43Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Targets a neglected class of defects that do not crash but corrupt training, providing a detailed protocol for reproducible debugging evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
