{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ETEZ4SW35A7M43SWO2DYEJQ4Y7","short_pith_number":"pith:ETEZ4SW3","schema_version":"1.0","canonical_sha256":"24c99e4adbe83ece6e56768782261cc7c5190a3ae8f5e34f53d5ae26582a8470","source":{"kind":"arxiv","id":"2310.05264","version":5},"attestation_state":"computed","paper":{"title":"The Emergence of Reproducibility and Generalizability in Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Huijie Zhang, Jinfan Zhou, Liyue Shen, Minzhe Guo, Peng Wang, Qing Qu, Yifu Lu","submitted_at":"2023-10-08T19:02:46Z","abstract_excerpt":"In this work, we investigate an intriguing and prevalent phenomenon of diffusion models which we term as \"consistent model reproducibility\": given the same starting noise input and a deterministic sampler, different diffusion models often yield remarkably similar outputs. We confirm this phenomenon through comprehensive experiments, implying that different diffusion models consistently reach the same data distribution and scoring function regardless of diffusion model frameworks, model architectures, or training procedures. More strikingly, our further investigation implies that diffusion mode"},"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":"2310.05264","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-08T19:02:46Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"b9745a2bf7825df4f4d408bd552e691d9eba20208c217fe7dbec129100f61d36","abstract_canon_sha256":"0a908a6e53302ae61a368114afa33eca3ba529f6d2bd5a1f03b3ae96b38ae69d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:09:10.931036Z","signature_b64":"wfR6gCqmG84AuBTvUOLHXYHK2Xh29pYeualpUmdzhxVmWDFk+U8lusxNQfa60FA+xs9/LpMj9V8HevdmSB6TCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24c99e4adbe83ece6e56768782261cc7c5190a3ae8f5e34f53d5ae26582a8470","last_reissued_at":"2026-06-10T01:09:10.929967Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:09:10.929967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Emergence of Reproducibility and Generalizability in Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Huijie Zhang, Jinfan Zhou, Liyue Shen, Minzhe Guo, Peng Wang, Qing Qu, Yifu Lu","submitted_at":"2023-10-08T19:02:46Z","abstract_excerpt":"In this work, we investigate an intriguing and prevalent phenomenon of diffusion models which we term as \"consistent model reproducibility\": given the same starting noise input and a deterministic sampler, different diffusion models often yield remarkably similar outputs. We confirm this phenomenon through comprehensive experiments, implying that different diffusion models consistently reach the same data distribution and scoring function regardless of diffusion model frameworks, model architectures, or training procedures. More strikingly, our further investigation implies that diffusion mode"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.05264","kind":"arxiv","version":5},"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/2310.05264/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":"2310.05264","created_at":"2026-06-10T01:09:10.930125+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.05264v5","created_at":"2026-06-10T01:09:10.930125+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.05264","created_at":"2026-06-10T01:09:10.930125+00:00"},{"alias_kind":"pith_short_12","alias_value":"ETEZ4SW35A7M","created_at":"2026-06-10T01:09:10.930125+00:00"},{"alias_kind":"pith_short_16","alias_value":"ETEZ4SW35A7M43SW","created_at":"2026-06-10T01:09:10.930125+00:00"},{"alias_kind":"pith_short_8","alias_value":"ETEZ4SW3","created_at":"2026-06-10T01:09:10.930125+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2602.22801","citing_title":"Unleashing the Potential of Diffusion Models for End-to-End Autonomous Driving","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2512.02826","citing_title":"From Navigation to Refinement: Revealing the Two-Stage Nature of Flow-based Diffusion Models through Oracle Velocity","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10019","citing_title":"The two clocks and the innovation window: When and how generative models learn rules","ref_index":84,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7","json":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7.json","graph_json":"https://pith.science/api/pith-number/ETEZ4SW35A7M43SWO2DYEJQ4Y7/graph.json","events_json":"https://pith.science/api/pith-number/ETEZ4SW35A7M43SWO2DYEJQ4Y7/events.json","paper":"https://pith.science/paper/ETEZ4SW3"},"agent_actions":{"view_html":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7","download_json":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7.json","view_paper":"https://pith.science/paper/ETEZ4SW3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.05264&json=true","fetch_graph":"https://pith.science/api/pith-number/ETEZ4SW35A7M43SWO2DYEJQ4Y7/graph.json","fetch_events":"https://pith.science/api/pith-number/ETEZ4SW35A7M43SWO2DYEJQ4Y7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7/action/storage_attestation","attest_author":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7/action/author_attestation","sign_citation":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7/action/citation_signature","submit_replication":"https://pith.science/pith/ETEZ4SW35A7M43SWO2DYEJQ4Y7/action/replication_record"}},"created_at":"2026-06-10T01:09:10.930125+00:00","updated_at":"2026-06-10T01:09:10.930125+00:00"}