{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.19881", "kind": "paper", "url": "https://arxiv.org/abs/2606.19881"}], "authors": ["Guneesh Vats", "Anubha Agrawal", "Shikha Singhal", "Ajita Dash", "Praison Selvaraj", "Vidhan Jhawar", "Ranga Prasad Chenna", "Bharadwaj Y M G"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.19881", "source_url": "https://arxiv.org/abs/2606.19881"}, "categories": ["General AI", "Vision & 3D", "cs.CL"], "description": {"en": "REDACT is a multilingual benchmark for PII detection with 13,427 records, 51 entity types, and controlled generation axes. It allows stratified evaluation via metadata fields and includes an evaluation harness."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.19881", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.19881", "title": "REDACT"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.19881", "modality": null, "name": "REDACT", "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_redact_293a3dff", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "20824b92e6ac1449bf5ea96bbddae800a51d9c0badebfcb49c581ba154d34bd3", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.19881"}, "publisher": null, "released": "2026-06-18", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.19881", "source_url": "https://arxiv.org/abs/2606.19881"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-19881", "source": "claire_radar", "source_benchmark_id": "2606.19881", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Mixed", "applicationDomains": ["General AI"], "area": "Vision & 3D", "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": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.19881", "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", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "stale"}, "capabilities": [], "confidence": "High", "construction": "Aggregate Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "b0ee0792c15da051396c72af01dd24996fd7c28a84d803ee769a11951ac804f7", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "REDACT is a multilingual benchmark for PII detection with 13,427 records, 51 entity types, and controlled generation axes. It allows stratified evaluation via metadata fields and includes an evaluation harness.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We present REDACT, a systematically controlled multilingual PII benchmark with 13,427 records, 324,078 entity annotations, 51 entity types, 4,127 surface-form patterns, and 25 languages across 9 scripts."}, "familyId": "bmf_18f1255e3f73", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_redact_293a3dff", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.19881", "paper": "https://arxiv.org/abs/2606.19881", "pdf": "https://arxiv.org/pdf/2606.19881", "project": null, "report": "https://arxiv.org/abs/2606.19881"}, "metrics": [], "motivation": "Benchmark infrastructure for personally identifiable information (PII) detection remains limited: existing corpora cover few entity types, use ad hoc generation conditions, and do not show which surface conditions cause detector failures.", "name": "REDACT", "oneLine": "REDACT is a multilingual benchmark for PII detection with 13,427 records, 51 entity types, and controlled generation axes. It allows stratified evaluation via metadata fields and includes an evaluation harness.", "paperTitle": "REDACT: A Systematically Controlled Multilingual Benchmark for Personal Information Detection", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.19881", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "REDACT Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2606.19881"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-18", "source": {"authors": ["Guneesh Vats", "Anubha Agrawal", "Shikha Singhal", "Ajita Dash", "Praison Selvaraj", "Vidhan Jhawar", "Ranga Prasad Chenna", "Bharadwaj Y M G"], "categories": ["cs.CL"], "id": "2606.19881", "title": "REDACT: A Systematically Controlled Multilingual Benchmark for Personal Information Detection", "type": "arxiv", "url": "https://arxiv.org/abs/2606.19881"}, "sourceUpdatedAt": "2026-06-18T07:38:15Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "PII detection lacks controlled benchmarks. REDACT provides systematic variation and layered evaluation to reveal failure conditions, aiding detector robustness."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
