{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.27472", "kind": "paper", "url": "https://arxiv.org/abs/2605.27472"}], "authors": ["Yuchao Wu", "Wenji Fang", "Jing Wang", "Wenkai Li", "Ziyan Guo", "Zhiyao Xie"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.27472", "source_url": "https://arxiv.org/abs/2605.27472"}, "categories": ["Chip Design & EDA", "Language & Knowledge", "cs.AR"], "description": {"en": "Evaluates LLM generation of SystemVerilog assertions from structured design specifications and RTL, across two tasks: bug-prevention and bug-hunting. The benchmark includes 83 real-world designs with golden and mutated RTL, and assesses syntactic validity, formal provability, coverage, and mutation-based bug detection."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.27472", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.27472", "title": "AssertLLM2"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Hong Kong University of Science and Technology"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2605.27472", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2605.27472", "source_url": "https://openalex.org/W7162686268"}, "key": "claire-radar:2605.27472", "modality": null, "name": "AssertLLM2", "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_assertllm2_9bd7c97e", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "15e95cabfbefd93dea5428cbbab51c98e1fbd36744db5512ee349e5d90e2b4e6", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.27472"}, "publisher": null, "released": "2026-05-26", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.27472", "source_url": "https://arxiv.org/abs/2605.27472"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-27472", "source": "claire_radar", "source_benchmark_id": "2605.27472", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Chip Design & EDA"], "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/2605.27472", "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": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "1fab1e04aac1f930b13d9a1b88636933d24d18d71520fe38d92588727e084c71", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates LLM generation of SystemVerilog assertions from structured design specifications and RTL, across two tasks: bug-prevention and bug-hunting. The benchmark includes 83 real-world designs with golden and mutated RTL, and assesses syntactic validity, formal provability, coverage, and mutation-based bug detection.", "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 these limitations, we introduce AssertLLM2, an open-source benchmark for realistic assertion generation in hardware verification."}, "familyId": "bmf_b6287b8c059f", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_assertllm2_9bd7c97e", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Semiconductors"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2605.27472", "paper": "https://arxiv.org/abs/2605.27472", "pdf": "https://arxiv.org/pdf/2605.27472", "project": null, "report": "https://arxiv.org/abs/2605.27472"}, "metrics": [], "motivation": "Assertion-based verification (ABV) is a cornerstone of modern hardware design, yet manually translating design intent into formal SystemVerilog Assertions (SVAs) remains labor-intensive and error-prone.", "name": "AssertLLM2", "oneLine": "Evaluates LLM generation of SystemVerilog assertions from structured design specifications and RTL, across two tasks: bug-prevention and bug-hunting. The benchmark includes 83 real-world designs with golden and mutated RTL, and assesses syntactic validity, formal provability, coverage, and mutation-based bug detection.", "paperTitle": "AssertLLM2: A Comprehensive LLM Benchmark for Assertion Generation from Design Specifications", "primaryDomain": "Chip Design & EDA", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.27472", "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": "2026-05-26", "source": {"authors": ["Yuchao Wu", "Wenji Fang", "Jing Wang", "Wenkai Li", "Ziyan Guo", "Zhiyao Xie"], "categories": ["cs.AR", "cs.AI"], "id": "2605.27472", "title": "AssertLLM2: A Comprehensive LLM Benchmark for Assertion Generation from Design Specifications", "type": "arxiv", "url": "https://arxiv.org/abs/2605.27472"}, "sourceUpdatedAt": "2026-05-26T10:49:25Z", "topics": ["cs.AR"], "venueAttempts": [], "whyItMatters": "Fills a gap in realistic evaluation for assertion generation by using full specifications and buggy RTL, supporting comparison of LLM capabilities for hardware verification tasks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
