{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.03983", "kind": "paper", "url": "https://arxiv.org/abs/2608.03983"}], "authors": ["Hailong Jiang", "Feng Yu", "Emran Hossain", "Jianfeng Zhu", "Mengfei Ren", "Qiang Guan", "Chunwei Xia"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.03983", "source_url": "https://arxiv.org/abs/2608.03983"}, "categories": ["Code & Software", "Software & AI Compute", "cs.PL"], "description": {"en": "SeGaBench is an executable benchmark containing 120 cases (100 synthetic, 20 source-backed) to test whether LLMs can recover semantic optimization opportunities that compilers miss, with hidden enabling semantics, oracle artifacts, and validators."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.03983", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.03983", "title": "SeGaBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Youngstown State University", "Kent State University", "Baylor University", "University of Leeds"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.03983", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.03983", "source_url": "https://openalex.org/W7172501395"}, "key": "claire-radar:2608.03983", "modality": null, "name": "SeGaBench", "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-indexed", "display_eligible": "unknown", "origin_record_id": "bm_segabench_e418e029", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "d32160e85520cebf50777820b975e61d380531c801353b62b13647d57cff3f07", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.03983"}, "publisher": null, "released": "2026-08-04", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.03983", "source_url": "https://arxiv.org/abs/2608.03983"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-03983", "source": "claire_radar", "source_benchmark_id": "2608.03983", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Machine Generated", "applicationDomains": ["Software & AI Compute"], "area": "Code & Software", "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/2608.03983", "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": "Medium", "construction": "Original Synthetic", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "d2b994507120a294e65116dbc46b3ccb8ae5d84c4912a6cf3e33f516c33fe3df", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "SeGaBench is an executable benchmark containing 120 cases (100 synthetic, 20 source-backed) to test whether LLMs can recover semantic optimization opportunities that compilers miss, with hidden enabling semantics, oracle artifacts, and validators.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence"], "snippet": "We introduce SeGaBench, an executable benchmark containing 100 synthetic and 20 source-backed cases spanning low-level assumptions, data-structure invariants, and high-level semantic lifting."}, "familyId": "bmf_7c6281aa1c49", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_segabench_e418e029", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.03983", "paper": "https://arxiv.org/abs/2608.03983", "pdf": "https://arxiv.org/pdf/2608.03983", "project": null, "report": "https://arxiv.org/abs/2608.03983"}, "metrics": [], "motivation": "Optimizing compilers miss profitable transformations when their enabling semantics are absent from the analyzed program representation.", "name": "SeGaBench", "oneLine": "SeGaBench is an executable benchmark containing 120 cases (100 synthetic, 20 source-backed) to test whether LLMs can recover semantic optimization opportunities that compilers miss, with hidden enabling semantics, oracle artifacts, and validators.", "paperTitle": "Can Large Language Models Recover Semantic Optimization Opportunities That Compilers Miss?", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.03983", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.0, "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": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}}, "readiness": "Paper only", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-08-04", "source": {"authors": ["Hailong Jiang", "Feng Yu", "Emran Hossain", "Jianfeng Zhu", "Mengfei Ren", "Qiang Guan", "Chunwei Xia"], "categories": ["cs.PL", "cs.AI"], "id": "2608.03983", "title": "Can Large Language Models Recover Semantic Optimization Opportunities That Compilers Miss?", "type": "arxiv", "url": "https://arxiv.org/abs/2608.03983"}, "sourceUpdatedAt": "2026-08-04T17:47:25Z", "topics": ["cs.PL"], "venueAttempts": [], "whyItMatters": "Compilers miss profitable transformations when enabling semantics are absent from the program representation. SeGaBench evaluates whether LLMs can recover such semantics and produce validated, performance-improving artifacts."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
