{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:25OCLWZQR6CALMSDS42EWX3PDE","short_pith_number":"pith:25OCLWZQ","schema_version":"1.0","canonical_sha256":"d75c25db308f8405b24397344b5f6f190d2d9d3d55b581c52bb30cf362714cfb","source":{"kind":"arxiv","id":"2607.19376","version":1},"attestation_state":"computed","paper":{"title":"Refnd: Preventing Data Leakage in Relational Datasets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.QM","authors_text":"Anthony Lavertu, Jacob Cote, Jacques Corbeil, Pascal Germain, Sophie Gobeil","submitted_at":"2026-06-30T19:53:33Z","abstract_excerpt":"Machine learning models trained on biochemical data are routinely evaluated using splits that fail to account for relational structure, causing information leakage and over-optimistic performance estimates. Existing splitting methods lack theoretical grounding and scale at best quadratically. We introduce the Relational Generative Process (RGP), a mathematical formalization explaining why relational structure arises in biochemical datasets, and Refnd, a splitting algorithm that leverages a proximity graph computed in loglinear time using Hierarchical Navigable Small World (HNSW). We validate o"},"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":"2607.19376","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"q-bio.QM","submitted_at":"2026-06-30T19:53:33Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"fb5f44cff87faae502258e68fa5cd53fd3103dbea16087fe607d7bd644bc115a","abstract_canon_sha256":"29fe74a7fbfc57116b78599c22067aba69490f88b0d9ca1f94cefa052391b9e0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-23T00:23:46.812844Z","signature_b64":"KfDm5QymbKpiAmaACyj2sExgMPGw6B3QJDY50YFaa3jZDJ48CxIzibJ/lnmmDoa09W33MBG4pLcL1YV8bqBrBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d75c25db308f8405b24397344b5f6f190d2d9d3d55b581c52bb30cf362714cfb","last_reissued_at":"2026-07-23T00:23:46.811972Z","signature_status":"signed_v1","first_computed_at":"2026-07-23T00:23:46.811972Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Refnd: Preventing Data Leakage in Relational Datasets","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.QM","authors_text":"Anthony Lavertu, Jacob Cote, Jacques Corbeil, Pascal Germain, Sophie Gobeil","submitted_at":"2026-06-30T19:53:33Z","abstract_excerpt":"Machine learning models trained on biochemical data are routinely evaluated using splits that fail to account for relational structure, causing information leakage and over-optimistic performance estimates. Existing splitting methods lack theoretical grounding and scale at best quadratically. We introduce the Relational Generative Process (RGP), a mathematical formalization explaining why relational structure arises in biochemical datasets, and Refnd, a splitting algorithm that leverages a proximity graph computed in loglinear time using Hierarchical Navigable Small World (HNSW). We validate o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.19376","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/2607.19376/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":"2607.19376","created_at":"2026-07-23T00:23:46.812425+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.19376v1","created_at":"2026-07-23T00:23:46.812425+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.19376","created_at":"2026-07-23T00:23:46.812425+00:00"},{"alias_kind":"pith_short_12","alias_value":"25OCLWZQR6CA","created_at":"2026-07-23T00:23:46.812425+00:00"},{"alias_kind":"pith_short_16","alias_value":"25OCLWZQR6CALMSD","created_at":"2026-07-23T00:23:46.812425+00:00"},{"alias_kind":"pith_short_8","alias_value":"25OCLWZQ","created_at":"2026-07-23T00:23:46.812425+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.08892","citing_title":"A Symmetric Layer-Union Audit of Component Collapse in Hierarchical Procedural Corpora","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE","json":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE.json","graph_json":"https://pith.science/api/pith-number/25OCLWZQR6CALMSDS42EWX3PDE/graph.json","events_json":"https://pith.science/api/pith-number/25OCLWZQR6CALMSDS42EWX3PDE/events.json","paper":"https://pith.science/paper/25OCLWZQ"},"agent_actions":{"view_html":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE","download_json":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE.json","view_paper":"https://pith.science/paper/25OCLWZQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.19376&json=true","fetch_graph":"https://pith.science/api/pith-number/25OCLWZQR6CALMSDS42EWX3PDE/graph.json","fetch_events":"https://pith.science/api/pith-number/25OCLWZQR6CALMSDS42EWX3PDE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE/action/storage_attestation","attest_author":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE/action/author_attestation","sign_citation":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE/action/citation_signature","submit_replication":"https://pith.science/pith/25OCLWZQR6CALMSDS42EWX3PDE/action/replication_record"}},"created_at":"2026-07-23T00:23:46.812425+00:00","updated_at":"2026-07-23T00:23:46.812425+00:00"}