{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:D4VJXTOGPBVFMKLNN6FSX77NOL","short_pith_number":"pith:D4VJXTOG","schema_version":"1.0","canonical_sha256":"1f2a9bcdc6786a56296d6f8b2bffed72d81a3b738a8ad2eca69836e992eece44","source":{"kind":"arxiv","id":"2402.06162","version":1},"attestation_state":"computed","paper":{"title":"Wasserstein proximal operators describe score-based generative models and resolve memorization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Benjamin J. Zhang, Markos A. Katsoulakis, Siting Liu, Stanley J. Osher, Wuchen Li","submitted_at":"2024-02-09T03:33:13Z","abstract_excerpt":"We focus on the fundamental mathematical structure of score-based generative models (SGMs). We first formulate SGMs in terms of the Wasserstein proximal operator (WPO) and demonstrate that, via mean-field games (MFGs), the WPO formulation reveals mathematical structure that describes the inductive bias of diffusion and score-based models. In particular, MFGs yield optimality conditions in the form of a pair of coupled partial differential equations: a forward-controlled Fokker-Planck (FP) equation, and a backward Hamilton-Jacobi-Bellman (HJB) equation. Via a Cole-Hopf transformation and taking"},"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":"2402.06162","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2024-02-09T03:33:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"327f09058f88bd1f646b89c655c73b433bdba4e274134d86d3305574f2e28e51","abstract_canon_sha256":"34a6361bf8c55d99df92f4aeeea5e1efd63f719cce707045a9b349052140780b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:43:19.689231Z","signature_b64":"6rokeRd31AcwzVn1Y8FwxPOg81xUHD+laphJv5dmheauTlqBulrAMECEOIpAcTox7nT366PC+U/EFXjJU6EjCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1f2a9bcdc6786a56296d6f8b2bffed72d81a3b738a8ad2eca69836e992eece44","last_reissued_at":"2026-07-05T07:43:19.688726Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:43:19.688726Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Wasserstein proximal operators describe score-based generative models and resolve memorization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Benjamin J. Zhang, Markos A. Katsoulakis, Siting Liu, Stanley J. Osher, Wuchen Li","submitted_at":"2024-02-09T03:33:13Z","abstract_excerpt":"We focus on the fundamental mathematical structure of score-based generative models (SGMs). We first formulate SGMs in terms of the Wasserstein proximal operator (WPO) and demonstrate that, via mean-field games (MFGs), the WPO formulation reveals mathematical structure that describes the inductive bias of diffusion and score-based models. In particular, MFGs yield optimality conditions in the form of a pair of coupled partial differential equations: a forward-controlled Fokker-Planck (FP) equation, and a backward Hamilton-Jacobi-Bellman (HJB) equation. Via a Cole-Hopf transformation and taking"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.06162","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/2402.06162/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":"2402.06162","created_at":"2026-07-05T07:43:19.688788+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.06162v1","created_at":"2026-07-05T07:43:19.688788+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.06162","created_at":"2026-07-05T07:43:19.688788+00:00"},{"alias_kind":"pith_short_12","alias_value":"D4VJXTOGPBVF","created_at":"2026-07-05T07:43:19.688788+00:00"},{"alias_kind":"pith_short_16","alias_value":"D4VJXTOGPBVFMKLN","created_at":"2026-07-05T07:43:19.688788+00:00"},{"alias_kind":"pith_short_8","alias_value":"D4VJXTOG","created_at":"2026-07-05T07:43:19.688788+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08953","citing_title":"Self-Consistent Generative Paths via Admissible Random Variational Transport","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2410.01244","citing_title":"Robustness and Structure Preservation in Flow-Based Generative Models via Wasserstein Path-Space Divergences","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10792","citing_title":"Implicit Neural Optimal Transport via Fixed-Point Optimization","ref_index":122,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL","json":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL.json","graph_json":"https://pith.science/api/pith-number/D4VJXTOGPBVFMKLNN6FSX77NOL/graph.json","events_json":"https://pith.science/api/pith-number/D4VJXTOGPBVFMKLNN6FSX77NOL/events.json","paper":"https://pith.science/paper/D4VJXTOG"},"agent_actions":{"view_html":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL","download_json":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL.json","view_paper":"https://pith.science/paper/D4VJXTOG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.06162&json=true","fetch_graph":"https://pith.science/api/pith-number/D4VJXTOGPBVFMKLNN6FSX77NOL/graph.json","fetch_events":"https://pith.science/api/pith-number/D4VJXTOGPBVFMKLNN6FSX77NOL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL/action/storage_attestation","attest_author":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL/action/author_attestation","sign_citation":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL/action/citation_signature","submit_replication":"https://pith.science/pith/D4VJXTOGPBVFMKLNN6FSX77NOL/action/replication_record"}},"created_at":"2026-07-05T07:43:19.688788+00:00","updated_at":"2026-07-05T07:43:19.688788+00:00"}