{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QJIDUJWIDOSRCL5BIJN2MCTV4W","short_pith_number":"pith:QJIDUJWI","schema_version":"1.0","canonical_sha256":"82503a26c81ba5112fa1425ba60a75e597221f855d3591eeb09f8021624bcb49","source":{"kind":"arxiv","id":"2604.15606","version":2},"attestation_state":"computed","paper":{"title":"Spec2Cov: An Agentic Framework for Code Coverage Closure of Digital Hardware Designs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Spec2Cov uses an LLM agent to generate and refine test stimulus from hardware design specifications, reaching full code coverage on simpler designs.","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Alma Babbit, Aman Arora, Elias Hilaneh, Nakul Gopalan, Sean Lowe, Vidya Chhabria","submitted_at":"2026-04-17T01:08:54Z","abstract_excerpt":"Hardware verification is one of the most challenging stages of the hardware design process, requiring significant time and resources to ensure a design is fully validated and production-ready. Verification teams aim to maximize design coverage while ensuring correct behavior and alignment with the specification. Coverage closure, which relies on iterative constrained-random and directed testing, is still largely manual and therefore slow and labor-intensive. Recent advances show that the code generation capabilities of Large Language Models (LLMs) can be integrated with external tools to build"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2604.15606","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AR","submitted_at":"2026-04-17T01:08:54Z","cross_cats_sorted":[],"title_canon_sha256":"76ab438e29371498b83ff82eb2ae82c2f54b6d0fc23bd80dfbe71048698760d3","abstract_canon_sha256":"31478f523e8a20c11f34e5fe868da56e1a3cac5f7a6b5e69190943f79058939d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T02:04:41.239301Z","signature_b64":"/ypLP+nbMYOQbXK0jhUg33mXrmlU5zoKK4qMgvmYGw0oaGa1sck9wSbXuqneL5wEdKUPph6XRCHDJ2vue50NCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"82503a26c81ba5112fa1425ba60a75e597221f855d3591eeb09f8021624bcb49","last_reissued_at":"2026-05-22T02:04:41.238382Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T02:04:41.238382Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spec2Cov: An Agentic Framework for Code Coverage Closure of Digital Hardware Designs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Spec2Cov uses an LLM agent to generate and refine test stimulus from hardware design specifications, reaching full code coverage on simpler designs.","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Alma Babbit, Aman Arora, Elias Hilaneh, Nakul Gopalan, Sean Lowe, Vidya Chhabria","submitted_at":"2026-04-17T01:08:54Z","abstract_excerpt":"Hardware verification is one of the most challenging stages of the hardware design process, requiring significant time and resources to ensure a design is fully validated and production-ready. Verification teams aim to maximize design coverage while ensuring correct behavior and alignment with the specification. Coverage closure, which relies on iterative constrained-random and directed testing, is still largely manual and therefore slow and labor-intensive. Recent advances show that the code generation capabilities of Large Language Models (LLMs) can be integrated with external tools to build"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across 26 designs of varying size and complexity, including problems from the CVDP benchmark suite, Spec2Cov demonstrates promising performance, achieving 100% coverage on simpler designs and up to 49% on more complex designs.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That an off-the-shelf LLM, guided only by coverage feedback and error messages, can produce stimulus that meaningfully advances coverage on realistic hardware without domain-specific fine-tuning or human repair of generated tests.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Spec2Cov uses an LLM-simulator feedback loop to generate tests from specs, reaching 100% coverage on simple designs and up to 49% on complex ones across 26 benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Spec2Cov uses an LLM agent to generate and refine test stimulus from hardware design specifications, reaching full code coverage on simpler designs.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"6cd7fd9c42345b49441bec7b41e45c3b957cbe2c6ec5310e24fa9095896f4e2b"},"source":{"id":"2604.15606","kind":"arxiv","version":2},"verdict":{"id":"8c2337db-f906-431f-ba58-ebf8fc370092","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T08:10:14.354877Z","strongest_claim":"Across 26 designs of varying size and complexity, including problems from the CVDP benchmark suite, Spec2Cov demonstrates promising performance, achieving 100% coverage on simpler designs and up to 49% on more complex designs.","one_line_summary":"Spec2Cov uses an LLM-simulator feedback loop to generate tests from specs, reaching 100% coverage on simple designs and up to 49% on complex ones across 26 benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That an off-the-shelf LLM, guided only by coverage feedback and error messages, can produce stimulus that meaningfully advances coverage on realistic hardware without domain-specific fine-tuning or human repair of generated tests.","pith_extraction_headline":"Spec2Cov uses an LLM agent to generate and refine test stimulus from hardware design specifications, reaching full code coverage on simpler designs."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.15606/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2604.15606","created_at":"2026-05-22T02:04:41.238548+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.15606v2","created_at":"2026-05-22T02:04:41.238548+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.15606","created_at":"2026-05-22T02:04:41.238548+00:00"},{"alias_kind":"pith_short_12","alias_value":"QJIDUJWIDOSR","created_at":"2026-05-22T02:04:41.238548+00:00"},{"alias_kind":"pith_short_16","alias_value":"QJIDUJWIDOSRCL5B","created_at":"2026-05-22T02:04:41.238548+00:00"},{"alias_kind":"pith_short_8","alias_value":"QJIDUJWI","created_at":"2026-05-22T02:04:41.238548+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W","json":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W.json","graph_json":"https://pith.science/api/pith-number/QJIDUJWIDOSRCL5BIJN2MCTV4W/graph.json","events_json":"https://pith.science/api/pith-number/QJIDUJWIDOSRCL5BIJN2MCTV4W/events.json","paper":"https://pith.science/paper/QJIDUJWI"},"agent_actions":{"view_html":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W","download_json":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W.json","view_paper":"https://pith.science/paper/QJIDUJWI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.15606&json=true","fetch_graph":"https://pith.science/api/pith-number/QJIDUJWIDOSRCL5BIJN2MCTV4W/graph.json","fetch_events":"https://pith.science/api/pith-number/QJIDUJWIDOSRCL5BIJN2MCTV4W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W/action/storage_attestation","attest_author":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W/action/author_attestation","sign_citation":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W/action/citation_signature","submit_replication":"https://pith.science/pith/QJIDUJWIDOSRCL5BIJN2MCTV4W/action/replication_record"}},"created_at":"2026-05-22T02:04:41.238548+00:00","updated_at":"2026-05-22T02:04:41.238548+00:00"}