{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.24765", "kind": "paper", "url": "https://arxiv.org/abs/2605.24765"}], "authors": ["Matilda Gaddi", "Jin Noh", "Onat Gungor", "Tajana Rosing"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.24765", "source_url": "https://arxiv.org/abs/2605.24765"}, "categories": ["Cybersecurity", "Language & Knowledge", "cs.CR"], "description": {"en": "The evaluation object is a dataset for privacy-aware cybersecurity question answering, covering key security domains with private entity labels. The main capability evaluated is QA accuracy and masking performance."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.24765", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.24765", "title": "CyberMaskQA"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of San Diego", "University of California San Diego"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2605.24765", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2605.24765", "source_url": "https://openalex.org/W7162400784"}, "key": "claire-radar:2605.24765", "modality": null, "name": "CyberMaskQA", "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_cybermaskqa_2283c85d", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "1dfd53a9e3f10ec8bb96f2061e57cf40a57a0aa3a1e68b42a16ccbe0d42c3b8e", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.24765"}, "publisher": null, "released": "2026-05-23", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.24765", "source_url": "https://arxiv.org/abs/2605.24765"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-24765", "source": "claire_radar", "source_benchmark_id": "2605.24765", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Cybersecurity"], "area": "Language & Knowledge", "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/2605.24765", "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": "62a9dc9cf905a20e3ecbddde28efeab2638bc9ec43694cb114944b77c4e58a21", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "The evaluation object is a dataset for privacy-aware cybersecurity question answering, covering key security domains with private entity labels. The main capability evaluated is QA accuracy and masking performance.", "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": "To address this gap, we introduce CYBERMASKQA, a privacy-aware QA benchmark covering key security domains."}, "familyId": "bmf_5ce23c89d555", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_cybermaskqa_2283c85d", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Cybersecurity"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2605.24765", "paper": "https://arxiv.org/abs/2605.24765", "pdf": "https://arxiv.org/pdf/2605.24765", "project": null, "report": "https://arxiv.org/abs/2605.24765"}, "metrics": [], "motivation": "Large language models (LLMs) are increasingly applied to cybersecurity question answering (QA) for critical tasks such as incident response and vulnerability analysis.", "name": "CyberMaskQA", "oneLine": "The evaluation object is a dataset for privacy-aware cybersecurity question answering, covering key security domains with private entity labels. The main capability evaluated is QA accuracy and masking performance.", "paperTitle": "CyberMaskQA: A Privacy-Aware Benchmark for Evaluating Large Language Models in Cybersecurity Question Answering", "primaryDomain": "Cybersecurity", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.24765", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-05-23", "source": {"authors": ["Matilda Gaddi", "Jin Noh", "Onat Gungor", "Tajana Rosing"], "categories": ["cs.CR", "cs.LG"], "id": "2605.24765", "title": "CyberMaskQA: A Privacy-Aware Benchmark for Evaluating Large Language Models in Cybersecurity Question Answering", "type": "arxiv", "url": "https://arxiv.org/abs/2605.24765"}, "sourceUpdatedAt": "2026-05-23T22:57:13Z", "topics": ["cs.CR"], "venueAttempts": [], "whyItMatters": "The evaluation gap is the lack of context-rich datasets for privacy-preserving QA in cybersecurity, which hinders progress. This benchmark's value is in enabling controlled information disclosure and studying privacy-utility trade-offs for deployable models."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
