{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T6HZKJVA3EY6LCAUFNXSPXT2W7","short_pith_number":"pith:T6HZKJVA","schema_version":"1.0","canonical_sha256":"9f8f9526a0d931e588142b6f27de7ab7d9bab5bbefbaa409c301343a4feb1d0c","source":{"kind":"arxiv","id":"2606.17252","version":1},"attestation_state":"computed","paper":{"title":"Stochastic Thermodynamics of Score Matching in Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.stat-mech"],"primary_cat":"cond-mat.dis-nn","authors_text":"H. T. Quan, Xuehao Ding, Yuhai Tu","submitted_at":"2026-06-15T19:54:50Z","abstract_excerpt":"Score-based diffusion models are a powerful class of generative AI systems capable of sampling from complex, high-dimensional probability distributions. Their dynamics consist of a forward diffusion process that transforms data into noise and a learned reverse process that reconstructs data by reversing the probability flow. Here, we develop a stochastic thermodynamic framework for diffusion models and their score-matching objective. We introduce a trajectory-dependent quantity, time-asymmetry entropy production (TAEP), defined from the forward and reverse diffusion dynamics, and show that it "},"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":"2606.17252","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cond-mat.dis-nn","submitted_at":"2026-06-15T19:54:50Z","cross_cats_sorted":["cond-mat.stat-mech"],"title_canon_sha256":"169afcff8db5f192688b890a90ab366a776589b82966eb072f8e7c4c9395cbe2","abstract_canon_sha256":"497882f593f5ca45fef6afeb52c0d3641a3c92e5a6fe18326fd1af2e8efb2630"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:06.938581Z","signature_b64":"rCt0hotkYDmNG52pPf9gSdeAey31T/t2JzJcSPbZracZ4L/n9yZ/8PK6OPCtc9kMcnfrKOit1xSa2h8UIIdTBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f8f9526a0d931e588142b6f27de7ab7d9bab5bbefbaa409c301343a4feb1d0c","last_reissued_at":"2026-06-19T16:10:06.938225Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:06.938225Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stochastic Thermodynamics of Score Matching in Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.stat-mech"],"primary_cat":"cond-mat.dis-nn","authors_text":"H. T. Quan, Xuehao Ding, Yuhai Tu","submitted_at":"2026-06-15T19:54:50Z","abstract_excerpt":"Score-based diffusion models are a powerful class of generative AI systems capable of sampling from complex, high-dimensional probability distributions. Their dynamics consist of a forward diffusion process that transforms data into noise and a learned reverse process that reconstructs data by reversing the probability flow. Here, we develop a stochastic thermodynamic framework for diffusion models and their score-matching objective. We introduce a trajectory-dependent quantity, time-asymmetry entropy production (TAEP), defined from the forward and reverse diffusion dynamics, and show that it "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.17252","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/2606.17252/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":"2606.17252","created_at":"2026-06-19T16:10:06.938285+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.17252v1","created_at":"2026-06-19T16:10:06.938285+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.17252","created_at":"2026-06-19T16:10:06.938285+00:00"},{"alias_kind":"pith_short_12","alias_value":"T6HZKJVA3EY6","created_at":"2026-06-19T16:10:06.938285+00:00"},{"alias_kind":"pith_short_16","alias_value":"T6HZKJVA3EY6LCAU","created_at":"2026-06-19T16:10:06.938285+00:00"},{"alias_kind":"pith_short_8","alias_value":"T6HZKJVA","created_at":"2026-06-19T16:10:06.938285+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.30451","citing_title":"Estimating Free Energy Differences with Virtually Escorted Trajectories","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7","json":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7.json","graph_json":"https://pith.science/api/pith-number/T6HZKJVA3EY6LCAUFNXSPXT2W7/graph.json","events_json":"https://pith.science/api/pith-number/T6HZKJVA3EY6LCAUFNXSPXT2W7/events.json","paper":"https://pith.science/paper/T6HZKJVA"},"agent_actions":{"view_html":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7","download_json":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7.json","view_paper":"https://pith.science/paper/T6HZKJVA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.17252&json=true","fetch_graph":"https://pith.science/api/pith-number/T6HZKJVA3EY6LCAUFNXSPXT2W7/graph.json","fetch_events":"https://pith.science/api/pith-number/T6HZKJVA3EY6LCAUFNXSPXT2W7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7/action/storage_attestation","attest_author":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7/action/author_attestation","sign_citation":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7/action/citation_signature","submit_replication":"https://pith.science/pith/T6HZKJVA3EY6LCAUFNXSPXT2W7/action/replication_record"}},"created_at":"2026-06-19T16:10:06.938285+00:00","updated_at":"2026-06-19T16:10:06.938285+00:00"}