{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:IO4ZGATAGEOMMWMKJPM66K3MN3","short_pith_number":"pith:IO4ZGATA","schema_version":"1.0","canonical_sha256":"43b9930260311cc6598a4bd9ef2b6c6ef83ecedebec6f3a4e3945a2bb960d4fc","source":{"kind":"arxiv","id":"2309.16467","version":2},"attestation_state":"computed","paper":{"title":"Compositional Program Generation for Few-Shot Systematic Generalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander Gray, Luke Liu, Maxwell Crouse, Parikshit Ram, Soham Dan, Tim Klinger","submitted_at":"2023-09-28T14:33:20Z","abstract_excerpt":"Compositional generalization is a key ability of humans that enables us to learn new concepts from only a handful examples. Neural machine learning models, including the now ubiquitous Transformers, struggle to generalize in this way, and typically require thousands of examples of a concept during training in order to generalize meaningfully. This difference in ability between humans and artificial neural architectures, motivates this study on a neuro-symbolic architecture called the Compositional Program Generator (CPG). CPG has three key features: \\textit{modularity}, \\textit{composition}, a"},"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":"2309.16467","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-09-28T14:33:20Z","cross_cats_sorted":[],"title_canon_sha256":"dcca36f7a08f5bc3d3cc46174fbc858855acffbc1bb5fd2b33fabbd56e0bb722","abstract_canon_sha256":"19e5ebd5c5be9d5a502075698d201a341c88bf45db191ddc9ae495fadc5d8dc1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:34:51.263216Z","signature_b64":"m9bT+CdzJpE2GzRmjTG8uReVynpSH4EwMvShbxPW0gY+rwPBcG/5Yne+0saayQeTkjn66QvKQgMXUJkb0i29CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43b9930260311cc6598a4bd9ef2b6c6ef83ecedebec6f3a4e3945a2bb960d4fc","last_reissued_at":"2026-07-05T07:34:51.262834Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:34:51.262834Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Compositional Program Generation for Few-Shot Systematic Generalization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander Gray, Luke Liu, Maxwell Crouse, Parikshit Ram, Soham Dan, Tim Klinger","submitted_at":"2023-09-28T14:33:20Z","abstract_excerpt":"Compositional generalization is a key ability of humans that enables us to learn new concepts from only a handful examples. Neural machine learning models, including the now ubiquitous Transformers, struggle to generalize in this way, and typically require thousands of examples of a concept during training in order to generalize meaningfully. This difference in ability between humans and artificial neural architectures, motivates this study on a neuro-symbolic architecture called the Compositional Program Generator (CPG). CPG has three key features: \\textit{modularity}, \\textit{composition}, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.16467","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/2309.16467/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":"2309.16467","created_at":"2026-07-05T07:34:51.262888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.16467v2","created_at":"2026-07-05T07:34:51.262888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.16467","created_at":"2026-07-05T07:34:51.262888+00:00"},{"alias_kind":"pith_short_12","alias_value":"IO4ZGATAGEOM","created_at":"2026-07-05T07:34:51.262888+00:00"},{"alias_kind":"pith_short_16","alias_value":"IO4ZGATAGEOMMWMK","created_at":"2026-07-05T07:34:51.262888+00:00"},{"alias_kind":"pith_short_8","alias_value":"IO4ZGATA","created_at":"2026-07-05T07:34:51.262888+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.03461","citing_title":"Solver-Aware Decompositions for Programming-by-Example: When Dividing Requires Knowing how to Conquer","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3","json":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3.json","graph_json":"https://pith.science/api/pith-number/IO4ZGATAGEOMMWMKJPM66K3MN3/graph.json","events_json":"https://pith.science/api/pith-number/IO4ZGATAGEOMMWMKJPM66K3MN3/events.json","paper":"https://pith.science/paper/IO4ZGATA"},"agent_actions":{"view_html":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3","download_json":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3.json","view_paper":"https://pith.science/paper/IO4ZGATA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.16467&json=true","fetch_graph":"https://pith.science/api/pith-number/IO4ZGATAGEOMMWMKJPM66K3MN3/graph.json","fetch_events":"https://pith.science/api/pith-number/IO4ZGATAGEOMMWMKJPM66K3MN3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3/action/storage_attestation","attest_author":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3/action/author_attestation","sign_citation":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3/action/citation_signature","submit_replication":"https://pith.science/pith/IO4ZGATAGEOMMWMKJPM66K3MN3/action/replication_record"}},"created_at":"2026-07-05T07:34:51.262888+00:00","updated_at":"2026-07-05T07:34:51.262888+00:00"}