{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2XPDFTZMVD2ITCJ2WJNAFKZFOU","short_pith_number":"pith:2XPDFTZM","schema_version":"1.0","canonical_sha256":"d5de32cf2ca8f489893ab25a02ab2575127323093674816571b1b4485513ea32","source":{"kind":"arxiv","id":"2607.27167","version":1},"attestation_state":"computed","paper":{"title":"SpecFirst: Behavioral Specification Elicitation as a First-Class Step in Agent-Based Program Synthesis from Scratch","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Ahmed E. Hassan, Boyuan Chen, Feng Lin, Khaled Chawa, Shaowei Wang, Shi Chang, Yihao Chen","submitted_at":"2026-07-29T17:42:47Z","abstract_excerpt":"LLM-based agents excel at software engineering tasks where an existing codebase provides context, but constructing a program from scratch remains fundamentally harder. Recent benchmarks such as ProgramBench quantify this gap: given only natural-language documentation and an execute-only binary as a behavioral oracle, even frontier models solve fewer than 1% of instances. Existing frameworks conflate documentation reading, behavioral exploration, and code synthesis into a single pass, causing agents to probe insufficiently, lose behavioral intent as context drifts, and propagate early misinterp"},"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":"2607.27167","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-07-29T17:42:47Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"e535b3093f82698f211081df36d313599f8029d80ef66559f9244357cc896142","abstract_canon_sha256":"b7c7f2b0108649e4c9f624f8cb7b48e33763da55d17ce17f53db47de3a7f413c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d5de32cf2ca8f489893ab25a02ab2575127323093674816571b1b4485513ea32","last_reissued_at":"2026-07-30T01:23:56.258292Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:23:56.258292Z"},"graph_snapshot":{"paper":{"title":"SpecFirst: Behavioral Specification Elicitation as a First-Class Step in Agent-Based Program Synthesis from Scratch","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Ahmed E. Hassan, Boyuan Chen, Feng Lin, Khaled Chawa, Shaowei Wang, Shi Chang, Yihao Chen","submitted_at":"2026-07-29T17:42:47Z","abstract_excerpt":"LLM-based agents excel at software engineering tasks where an existing codebase provides context, but constructing a program from scratch remains fundamentally harder. Recent benchmarks such as ProgramBench quantify this gap: given only natural-language documentation and an execute-only binary as a behavioral oracle, even frontier models solve fewer than 1% of instances. Existing frameworks conflate documentation reading, behavioral exploration, and code synthesis into a single pass, causing agents to probe insufficiently, lose behavioral intent as context drifts, and propagate early misinterp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27167","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/2607.27167/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":"2607.27167","created_at":"2026-07-30T01:23:56.263280+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.27167v1","created_at":"2026-07-30T01:23:56.263280+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27167","created_at":"2026-07-30T01:23:56.263280+00:00"},{"alias_kind":"pith_short_12","alias_value":"2XPDFTZMVD2I","created_at":"2026-07-30T01:23:56.263280+00:00"},{"alias_kind":"pith_short_16","alias_value":"2XPDFTZMVD2ITCJ2","created_at":"2026-07-30T01:23:56.263280+00:00"},{"alias_kind":"pith_short_8","alias_value":"2XPDFTZM","created_at":"2026-07-30T01:23:56.263280+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/2XPDFTZMVD2ITCJ2WJNAFKZFOU","json":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU.json","graph_json":"https://pith.science/api/pith-number/2XPDFTZMVD2ITCJ2WJNAFKZFOU/graph.json","events_json":"https://pith.science/api/pith-number/2XPDFTZMVD2ITCJ2WJNAFKZFOU/events.json","paper":"https://pith.science/paper/2XPDFTZM"},"agent_actions":{"view_html":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU","download_json":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU.json","view_paper":"https://pith.science/paper/2XPDFTZM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.27167&json=true","fetch_graph":"https://pith.science/api/pith-number/2XPDFTZMVD2ITCJ2WJNAFKZFOU/graph.json","fetch_events":"https://pith.science/api/pith-number/2XPDFTZMVD2ITCJ2WJNAFKZFOU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU/action/storage_attestation","attest_author":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU/action/author_attestation","sign_citation":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU/action/citation_signature","submit_replication":"https://pith.science/pith/2XPDFTZMVD2ITCJ2WJNAFKZFOU/action/replication_record"}},"created_at":"2026-07-30T01:23:56.263280+00:00","updated_at":"2026-07-30T01:23:56.263280+00:00"}