{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3ERMXYEUNL7GILRH3LOW5IB4EF","short_pith_number":"pith:3ERMXYEU","schema_version":"1.0","canonical_sha256":"d922cbe0946afe642e27dadd6ea03c2150661965adb300971c748d37d3e3e814","source":{"kind":"arxiv","id":"2605.30314","version":1},"attestation_state":"computed","paper":{"title":"SpecBench: Evaluating Specification-Level Reasoning for Software Engineering LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.MA","authors_text":"Anand Jayarajan, Gennady Pekhimenko, Grant Hamblin, Kevin Song, Nandita Vijaykumar, Sihang Liu, Zhanda Zhu","submitted_at":"2026-05-28T17:54:01Z","abstract_excerpt":"Software engineering (SWE) agents are transitioning from code generation to full software development lifecycle automation. A critical phase in this lifecycle is specification design: transforming initial proposals into carefully considered requirements through expert review. Existing benchmarks such as SWE-Bench are implementation-focused by measuring the agent's ability to generate code given fixed, precise design requirements. This formulation assumes specifications are correct and complete. In real-world complex and critical software systems, initial specifications are often incomplete and"},"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":"2605.30314","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MA","submitted_at":"2026-05-28T17:54:01Z","cross_cats_sorted":[],"title_canon_sha256":"9c4a98552721c5dcdddca6b5c5e4965b86619c84403e8417b1dc013a5ab5bd08","abstract_canon_sha256":"776be9b3b336d60fb629c8e38997a3ccbad529873805d864e7bafc74cce5657b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T02:06:16.233942Z","signature_b64":"UTkl8o9z4mDT7OO5mKg+73+0ud5xGTaWRegizz8GV+DUCfSCLotBO5I44mdgolr78NC6m3KqstYYK4z2U0XdDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d922cbe0946afe642e27dadd6ea03c2150661965adb300971c748d37d3e3e814","last_reissued_at":"2026-05-29T02:06:16.233553Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T02:06:16.233553Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SpecBench: Evaluating Specification-Level Reasoning for Software Engineering LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.MA","authors_text":"Anand Jayarajan, Gennady Pekhimenko, Grant Hamblin, Kevin Song, Nandita Vijaykumar, Sihang Liu, Zhanda Zhu","submitted_at":"2026-05-28T17:54:01Z","abstract_excerpt":"Software engineering (SWE) agents are transitioning from code generation to full software development lifecycle automation. A critical phase in this lifecycle is specification design: transforming initial proposals into carefully considered requirements through expert review. Existing benchmarks such as SWE-Bench are implementation-focused by measuring the agent's ability to generate code given fixed, precise design requirements. This formulation assumes specifications are correct and complete. In real-world complex and critical software systems, initial specifications are often incomplete and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30314","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.30314/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":"2605.30314","created_at":"2026-05-29T02:06:16.233620+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30314v1","created_at":"2026-05-29T02:06:16.233620+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30314","created_at":"2026-05-29T02:06:16.233620+00:00"},{"alias_kind":"pith_short_12","alias_value":"3ERMXYEUNL7G","created_at":"2026-05-29T02:06:16.233620+00:00"},{"alias_kind":"pith_short_16","alias_value":"3ERMXYEUNL7GILRH","created_at":"2026-05-29T02:06:16.233620+00:00"},{"alias_kind":"pith_short_8","alias_value":"3ERMXYEU","created_at":"2026-05-29T02:06:16.233620+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/3ERMXYEUNL7GILRH3LOW5IB4EF","json":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF.json","graph_json":"https://pith.science/api/pith-number/3ERMXYEUNL7GILRH3LOW5IB4EF/graph.json","events_json":"https://pith.science/api/pith-number/3ERMXYEUNL7GILRH3LOW5IB4EF/events.json","paper":"https://pith.science/paper/3ERMXYEU"},"agent_actions":{"view_html":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF","download_json":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF.json","view_paper":"https://pith.science/paper/3ERMXYEU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30314&json=true","fetch_graph":"https://pith.science/api/pith-number/3ERMXYEUNL7GILRH3LOW5IB4EF/graph.json","fetch_events":"https://pith.science/api/pith-number/3ERMXYEUNL7GILRH3LOW5IB4EF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF/action/storage_attestation","attest_author":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF/action/author_attestation","sign_citation":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF/action/citation_signature","submit_replication":"https://pith.science/pith/3ERMXYEUNL7GILRH3LOW5IB4EF/action/replication_record"}},"created_at":"2026-05-29T02:06:16.233620+00:00","updated_at":"2026-05-29T02:06:16.233620+00:00"}