{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.13522", "kind": "paper", "url": "https://arxiv.org/abs/2608.13522"}, {"id": "gh:sunblaze-ucb/vero", "kind": "repo", "url": "https://github.com/sunblaze-ucb/vero"}], "authors": ["Zhe Ye", "Hantao Lou", "Yuechun Sun", "Peiyang Song", "Zhengxu Yan", "Timothe Kasriel", "Qingyang Zhang", "Kaiyu Yang", "Soonho Kong", "Jingxuan He", "Dawn Song"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.13522", "source_url": "https://arxiv.org/abs/2608.13522"}, "categories": ["Language & Knowledge", "Software & AI Compute", "cs.LG"], "description": {"en": "Vero evaluates repository-level verified code generation in Lean 4, with 43 multi-module instances, formal specifications, and proof-only and code-and-proof evaluation modes."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.13522", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.13522", "title": "Vero"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of California, Berkeley", "University of Chicago", "California Institute of Technology", "Stanford University", "Amazon (United States)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.13522", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.13522", "source_url": "https://openalex.org/W7203470636"}, "key": "claire-radar:2608.13522", "modality": null, "name": "Vero", "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_vero_3ed8d572", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "87e0980a340fb8bec85838e512b4b5077e9f374cd09a3ce832da8ff481cadb51", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.13522"}, "publisher": null, "released": "2026-08-13", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.13522", "source_url": "https://arxiv.org/abs/2608.13522"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-13522", "source": "claire_radar", "source_benchmark_id": "2608.13522", "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/sunblaze-ucb/vero", "githubScope": "benchmark_repo", "githubStars": 105, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2608.13522", "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": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-30", "state": "stale"}}, "status": "fresh"}, "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": "388e10a5152ca39158cc75b598822a9c50dee75c9bfe67498740462ee46d4da5", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Vero evaluates repository-level verified code generation in Lean 4, with 43 multi-module instances, formal specifications, and proof-only and code-and-proof evaluation modes.", "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", "public artifact URL"], "snippet": "To bridge this gap, we introduce Vero, the first benchmark to evaluate joint implementation and proof synthesis at the repository level."}, "familyId": "bmf_5cca655ac072", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/sunblaze-ucb/vero", "url": "https://github.com/sunblaze-ucb/vero"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/sunblaze-ucb/vero", "url": "https://huggingface.co/papers/2608.13522"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/sunblaze-ucb/vero.", "url": "https://arxiv.org/abs/2608.13522"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/sunblaze-ucb/vero", "url": "https://arxiv.org/abs/2608.13522"}], "scope": "project_repo", "url": "https://github.com/sunblaze-ucb/vero", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_vero_3ed8d572", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Software & Cloud"], "links": {"code": "https://github.com/sunblaze-ucb/vero", "data": null, "hfPaper": "https://huggingface.co/papers/2608.13522", "paper": "https://arxiv.org/abs/2608.13522", "pdf": "https://arxiv.org/pdf/2608.13522", "project": null, "report": "https://arxiv.org/abs/2608.13522"}, "metrics": [], "motivation": "AI agents are increasingly used for programming, but do not provide any guarantee on the correctness of generated code.", "name": "Vero", "oneLine": "Vero evaluates repository-level verified code generation in Lean 4, with 43 multi-module instances, formal specifications, and proof-only and code-and-proof evaluation modes.", "paperTitle": "Vero: Can AI Agents Build Formally Verified Software Repositories?", "primaryDomain": "Software & AI Compute", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.13522", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Sunblaze UCB", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/sunblaze-ucb/vero"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "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": 0.6224852885180543, "value": 105}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.0, "value": 0}}, "confidence": "Medium", "coverage": 0.7, "rank": 70, "score": 49}, "method": "attention-ranking-v13", "rank": 70, "score": 49}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-13", "source": {"authors": ["Zhe Ye", "Hantao Lou", "Yuechun Sun", "Peiyang Song", "Zhengxu Yan", "Timothe Kasriel", "Qingyang Zhang", "Kaiyu Yang", "Soonho Kong", "Jingxuan He", "Dawn Song"], "categories": ["cs.LG", "cs.AI", "cs.LO", "cs.PL", "cs.SE"], "id": "2608.13522", "title": "Vero: Can AI Agents Build Formally Verified Software Repositories?", "type": "arxiv", "url": "https://arxiv.org/abs/2608.13522"}, "sourceUpdatedAt": "2026-08-13T17:41:27Z", "topics": ["cs.LG"], "venueAttempts": [], "whyItMatters": "It provides a testbed for measuring progress toward repository-scale verified software synthesis, where current agents fall short."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
