{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:O64Z5DXFCCYDTL6RLSU3YRJ4RZ","short_pith_number":"pith:O64Z5DXF","schema_version":"1.0","canonical_sha256":"77b99e8ee510b039afd15ca9bc453c8e59cafa6cd657f5368468f8a364061d0c","source":{"kind":"arxiv","id":"2507.07855","version":4},"attestation_state":"computed","paper":{"title":"DPO Unchained: Your Training Algorithm is Secretly Disentangled in Human Choice Theory (and its Loss' Convexity is Dispensable)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Andrew Hard, Brice Magdalou, Ehsan Amid, John Lambert, Richard Nock, Shujian Zhang, Wenxuan Zhou","submitted_at":"2025-07-10T15:38:17Z","abstract_excerpt":"Normative theories allow one to elicit key parts of a ML algorithm from first principles, which is crucial at a time of championed scrutiny for ML work. Direct Preference Optimization (DPO) cleverly bypasses reward modeling by making an explicit link with a specific normative model of human choice. Our paper elevates this connection to the full generality of DPO's normative framework. Getting there requires reworking human choice theory's textbook path for a better RLHF/ML fit. It elevates the connection to a remarkably broad viewpoint on preference optimization, considering the current panora"},"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":"2507.07855","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-10T15:38:17Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"133a594afe60225e8a07f2b91e48d218d330ccf816a2155f6375582126dc4fdd","abstract_canon_sha256":"4742bb624bb30ffac47b6b4e06ca003eb230615cd0d6baa30665c020c99f4e88"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T03:13:46.185258Z","signature_b64":"hDlTxL3B8S8dx2NgD4zsk/njBJsF1xlHy9mz7u/SD1HjVAZe138RRX8YYtMwv7E5aBl73yKc47cqwj7q2cskAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"77b99e8ee510b039afd15ca9bc453c8e59cafa6cd657f5368468f8a364061d0c","last_reissued_at":"2026-06-23T03:13:46.184638Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T03:13:46.184638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DPO Unchained: Your Training Algorithm is Secretly Disentangled in Human Choice Theory (and its Loss' Convexity is Dispensable)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Andrew Hard, Brice Magdalou, Ehsan Amid, John Lambert, Richard Nock, Shujian Zhang, Wenxuan Zhou","submitted_at":"2025-07-10T15:38:17Z","abstract_excerpt":"Normative theories allow one to elicit key parts of a ML algorithm from first principles, which is crucial at a time of championed scrutiny for ML work. Direct Preference Optimization (DPO) cleverly bypasses reward modeling by making an explicit link with a specific normative model of human choice. Our paper elevates this connection to the full generality of DPO's normative framework. Getting there requires reworking human choice theory's textbook path for a better RLHF/ML fit. It elevates the connection to a remarkably broad viewpoint on preference optimization, considering the current panora"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07855","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/2507.07855/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":"2507.07855","created_at":"2026-06-23T03:13:46.184723+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.07855v4","created_at":"2026-06-23T03:13:46.184723+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07855","created_at":"2026-06-23T03:13:46.184723+00:00"},{"alias_kind":"pith_short_12","alias_value":"O64Z5DXFCCYD","created_at":"2026-06-23T03:13:46.184723+00:00"},{"alias_kind":"pith_short_16","alias_value":"O64Z5DXFCCYDTL6R","created_at":"2026-06-23T03:13:46.184723+00:00"},{"alias_kind":"pith_short_8","alias_value":"O64Z5DXF","created_at":"2026-06-23T03:13:46.184723+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/O64Z5DXFCCYDTL6RLSU3YRJ4RZ","json":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ.json","graph_json":"https://pith.science/api/pith-number/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/graph.json","events_json":"https://pith.science/api/pith-number/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/events.json","paper":"https://pith.science/paper/O64Z5DXF"},"agent_actions":{"view_html":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ","download_json":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ.json","view_paper":"https://pith.science/paper/O64Z5DXF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.07855&json=true","fetch_graph":"https://pith.science/api/pith-number/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/graph.json","fetch_events":"https://pith.science/api/pith-number/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/action/storage_attestation","attest_author":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/action/author_attestation","sign_citation":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/action/citation_signature","submit_replication":"https://pith.science/pith/O64Z5DXFCCYDTL6RLSU3YRJ4RZ/action/replication_record"}},"created_at":"2026-06-23T03:13:46.184723+00:00","updated_at":"2026-06-23T03:13:46.184723+00:00"}