{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CQAMZETIMWHKKSVH44N4VVBZ2R","short_pith_number":"pith:CQAMZETI","schema_version":"1.0","canonical_sha256":"1400cc9268658ea54aa7e71bcad439d45724d1d1559b1ccd85ddceb36b139308","source":{"kind":"arxiv","id":"2401.16467","version":2},"attestation_state":"computed","paper":{"title":"ReGAL: Refactoring Programs to Discover Generalizable Abstractions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.PL"],"primary_cat":"cs.SE","authors_text":"Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal","submitted_at":"2024-01-29T18:45:30Z","abstract_excerpt":"While large language models (LLMs) are increasingly being used for program synthesis, they lack the global view needed to develop useful abstractions; they generally predict programs one at a time, often repeating the same functionality. Generating redundant code from scratch is both inefficient and error-prone. To address this, we propose Refactoring for Generalizable Abstraction Learning (ReGAL), a gradient-free method for learning a library of reusable functions via code refactorization, i.e., restructuring code without changing its execution output. ReGAL learns from a small set of existin"},"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":"2401.16467","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-01-29T18:45:30Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG","cs.PL"],"title_canon_sha256":"01752d42ad20263eff8136c728f32444109a7d344225c95651a0b8d761c78ef5","abstract_canon_sha256":"a9790a77a73fa8430d947b388c92a4a39f0787746389339eb7482afd6df47cd7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:26.867753Z","signature_b64":"ZT52nGRNviS2qCO9OB2DvVxQvXUmcKx+dRhPgxnOcRB/HuTAenZRUR4EVPk3ZKmU6qDABJu5MW9twhFfTOM2Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1400cc9268658ea54aa7e71bcad439d45724d1d1559b1ccd85ddceb36b139308","last_reissued_at":"2026-07-05T08:28:26.867356Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:26.867356Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ReGAL: Refactoring Programs to Discover Generalizable Abstractions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG","cs.PL"],"primary_cat":"cs.SE","authors_text":"Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal","submitted_at":"2024-01-29T18:45:30Z","abstract_excerpt":"While large language models (LLMs) are increasingly being used for program synthesis, they lack the global view needed to develop useful abstractions; they generally predict programs one at a time, often repeating the same functionality. Generating redundant code from scratch is both inefficient and error-prone. To address this, we propose Refactoring for Generalizable Abstraction Learning (ReGAL), a gradient-free method for learning a library of reusable functions via code refactorization, i.e., restructuring code without changing its execution output. ReGAL learns from a small set of existin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.16467","kind":"arxiv","version":2},"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/2401.16467/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":"2401.16467","created_at":"2026-07-05T08:28:26.867414+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.16467v2","created_at":"2026-07-05T08:28:26.867414+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.16467","created_at":"2026-07-05T08:28:26.867414+00:00"},{"alias_kind":"pith_short_12","alias_value":"CQAMZETIMWHK","created_at":"2026-07-05T08:28:26.867414+00:00"},{"alias_kind":"pith_short_16","alias_value":"CQAMZETIMWHKKSVH","created_at":"2026-07-05T08:28:26.867414+00:00"},{"alias_kind":"pith_short_8","alias_value":"CQAMZETI","created_at":"2026-07-05T08:28:26.867414+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.16046","citing_title":"XSearch: Explainable Code Search via Concept-to-Code Alignment","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16046","citing_title":"XSearch: Explainable Code Search via Concept-to-Code Alignment","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08468","citing_title":"PYTHALAB-MERA: Validation-Grounded Memory, Retrieval, and Acceptance Control for Frozen-LLM Coding Agents","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05485","citing_title":"ReaComp: Compiling LLM Reasoning into Symbolic Solvers for Efficient Program Synthesis","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20759","citing_title":"Autark: A Serverless Toolkit for Prototyping Urban Visual Analytics Systems","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20759","citing_title":"Autark: A Serverless Toolkit for Prototyping Urban Visual Analytics Systems","ref_index":58,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R","json":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R.json","graph_json":"https://pith.science/api/pith-number/CQAMZETIMWHKKSVH44N4VVBZ2R/graph.json","events_json":"https://pith.science/api/pith-number/CQAMZETIMWHKKSVH44N4VVBZ2R/events.json","paper":"https://pith.science/paper/CQAMZETI"},"agent_actions":{"view_html":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R","download_json":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R.json","view_paper":"https://pith.science/paper/CQAMZETI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.16467&json=true","fetch_graph":"https://pith.science/api/pith-number/CQAMZETIMWHKKSVH44N4VVBZ2R/graph.json","fetch_events":"https://pith.science/api/pith-number/CQAMZETIMWHKKSVH44N4VVBZ2R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R/action/storage_attestation","attest_author":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R/action/author_attestation","sign_citation":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R/action/citation_signature","submit_replication":"https://pith.science/pith/CQAMZETIMWHKKSVH44N4VVBZ2R/action/replication_record"}},"created_at":"2026-07-05T08:28:26.867414+00:00","updated_at":"2026-07-05T08:28:26.867414+00:00"}