{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JO4ZQP6ZNOF2IVXQ5JWCTAJGH2","short_pith_number":"pith:JO4ZQP6Z","schema_version":"1.0","canonical_sha256":"4bb9983fd96b8ba456f0ea6c2981263ebc7c3c87f668b03d4da4bfd8a5b8818d","source":{"kind":"arxiv","id":"2408.02232","version":4},"attestation_state":"computed","paper":{"title":"SpecRover: Code Intent Extraction via LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Abhik Roychoudhury, Haifeng Ruan, Yuntong Zhang","submitted_at":"2024-08-05T04:53:01Z","abstract_excerpt":"Autonomous program improvement typically involves automatically producing bug fixes and feature additions. Such program improvement can be accomplished by a combination of large language model (LLM) and program analysis capabilities, in the form of an LLM agent. Since program repair or program improvement typically requires a specification of intended behavior - specification inference can be useful for producing high quality program patches. In this work, we examine efficient and low-cost workflows for iterative specification inference within an LLM agent. Given a GitHub issue to be resolved "},"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":"2408.02232","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-08-05T04:53:01Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"65a203bc3071ffae1e32b790722f97cb854503f13abc78e7931b8815401c9207","abstract_canon_sha256":"c49bc738b501740799c0bf96e8be4636d6105620f2fbf148d575774e192f81ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:47:32.980981Z","signature_b64":"9kiiqXWYzcgN5cMSYtiC7neEOrVYofLijBu6HgfmpHrO5cHIpBq7xmj9oXx0LNgh8lpHwp6xam92aVkb39IaBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4bb9983fd96b8ba456f0ea6c2981263ebc7c3c87f668b03d4da4bfd8a5b8818d","last_reissued_at":"2026-07-05T09:47:32.980443Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:47:32.980443Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SpecRover: Code Intent Extraction via LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Abhik Roychoudhury, Haifeng Ruan, Yuntong Zhang","submitted_at":"2024-08-05T04:53:01Z","abstract_excerpt":"Autonomous program improvement typically involves automatically producing bug fixes and feature additions. Such program improvement can be accomplished by a combination of large language model (LLM) and program analysis capabilities, in the form of an LLM agent. Since program repair or program improvement typically requires a specification of intended behavior - specification inference can be useful for producing high quality program patches. In this work, we examine efficient and low-cost workflows for iterative specification inference within an LLM agent. Given a GitHub issue to be resolved "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.02232","kind":"arxiv","version":4},"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/2408.02232/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":"2408.02232","created_at":"2026-07-05T09:47:32.980503+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.02232v4","created_at":"2026-07-05T09:47:32.980503+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.02232","created_at":"2026-07-05T09:47:32.980503+00:00"},{"alias_kind":"pith_short_12","alias_value":"JO4ZQP6ZNOF2","created_at":"2026-07-05T09:47:32.980503+00:00"},{"alias_kind":"pith_short_16","alias_value":"JO4ZQP6ZNOF2IVXQ","created_at":"2026-07-05T09:47:32.980503+00:00"},{"alias_kind":"pith_short_8","alias_value":"JO4ZQP6Z","created_at":"2026-07-05T09:47:32.980503+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2503.12374","citing_title":"Beyond Final Code: A Process-Oriented Error Analysis of Software Development Agents in Real-World GitHub Scenarios","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2511.16858","citing_title":"Investigating Test Overfitting on SWE-bench","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2602.07900","citing_title":"Rethinking the Value of Agent-Generated Tests for LLM-Based Software Engineering Agents","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2603.22048","citing_title":"Dynamic analysis enhances issue resolution","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2407.01489","citing_title":"Agentless: Demystifying LLM-based Software Engineering Agents","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12108","citing_title":"LLM-Based Automated Diagnosis Of Integration Test Failures At Google","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04580","citing_title":"Beyond Fixed Tests: Repository-Level Issue Resolution as Coevolution of Code and Behavioral Constraints","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2","json":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2.json","graph_json":"https://pith.science/api/pith-number/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/graph.json","events_json":"https://pith.science/api/pith-number/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/events.json","paper":"https://pith.science/paper/JO4ZQP6Z"},"agent_actions":{"view_html":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2","download_json":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2.json","view_paper":"https://pith.science/paper/JO4ZQP6Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.02232&json=true","fetch_graph":"https://pith.science/api/pith-number/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/graph.json","fetch_events":"https://pith.science/api/pith-number/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/action/storage_attestation","attest_author":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/action/author_attestation","sign_citation":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/action/citation_signature","submit_replication":"https://pith.science/pith/JO4ZQP6ZNOF2IVXQ5JWCTAJGH2/action/replication_record"}},"created_at":"2026-07-05T09:47:32.980503+00:00","updated_at":"2026-07-05T09:47:32.980503+00:00"}