{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SY5N2IVHSL2LHVA4JLUOBGVSWJ","short_pith_number":"pith:SY5N2IVH","schema_version":"1.0","canonical_sha256":"963add22a792f4b3d41c4ae8e09ab2b24084152f3a38644a6e8283c561a95f7c","source":{"kind":"arxiv","id":"2605.24192","version":1},"attestation_state":"computed","paper":{"title":"Filtered Posterior Mean Collections: A Unified Framework for Analytical Models of Diffusion Generalization","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Berend Zwartsenberg, Frank Wood, Matthew Niedoba","submitted_at":"2026-05-22T20:29:04Z","abstract_excerpt":"The neural-network denoising functions which form the backbone of image diffusion models are remarkably consistent in their generalization behaviour across a wide variety of network architectures and training procedure hyperparameters. A recent line of research has sought to model the outputs of these networks by aggregating posterior weighted averages of training dataset patches. In this work, we consolidate these approaches into a unified model class which we call Filtered Posterior Mean Collections (FPMCs). We define this model class using query precision vectors, response weights, and sour"},"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":"2605.24192","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-22T20:29:04Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"209156bf55925c6bd4e55048f7a6f8fd6fd6e3295a6b4d2ee0d747a5f98b1bb1","abstract_canon_sha256":"03487a2fe76df5337fe0792d3dd1248a6f2b178080d6eec7d9f40436550874b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:02:51.707410Z","signature_b64":"ht0YxFJunHX8prPCwlFpnLqNAU/Xr5SAvQ96Vmod9ccp/6Jf5u5bryNl4JzfXag0wkPdS70MNHRPGzRJ4ilFCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"963add22a792f4b3d41c4ae8e09ab2b24084152f3a38644a6e8283c561a95f7c","last_reissued_at":"2026-05-26T01:02:51.706638Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:02:51.706638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Filtered Posterior Mean Collections: A Unified Framework for Analytical Models of Diffusion Generalization","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Berend Zwartsenberg, Frank Wood, Matthew Niedoba","submitted_at":"2026-05-22T20:29:04Z","abstract_excerpt":"The neural-network denoising functions which form the backbone of image diffusion models are remarkably consistent in their generalization behaviour across a wide variety of network architectures and training procedure hyperparameters. A recent line of research has sought to model the outputs of these networks by aggregating posterior weighted averages of training dataset patches. In this work, we consolidate these approaches into a unified model class which we call Filtered Posterior Mean Collections (FPMCs). We define this model class using query precision vectors, response weights, and sour"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24192","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/2605.24192/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":"2605.24192","created_at":"2026-05-26T01:02:51.706755+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24192v1","created_at":"2026-05-26T01:02:51.706755+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24192","created_at":"2026-05-26T01:02:51.706755+00:00"},{"alias_kind":"pith_short_12","alias_value":"SY5N2IVHSL2L","created_at":"2026-05-26T01:02:51.706755+00:00"},{"alias_kind":"pith_short_16","alias_value":"SY5N2IVHSL2LHVA4","created_at":"2026-05-26T01:02:51.706755+00:00"},{"alias_kind":"pith_short_8","alias_value":"SY5N2IVH","created_at":"2026-05-26T01:02:51.706755+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/SY5N2IVHSL2LHVA4JLUOBGVSWJ","json":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ.json","graph_json":"https://pith.science/api/pith-number/SY5N2IVHSL2LHVA4JLUOBGVSWJ/graph.json","events_json":"https://pith.science/api/pith-number/SY5N2IVHSL2LHVA4JLUOBGVSWJ/events.json","paper":"https://pith.science/paper/SY5N2IVH"},"agent_actions":{"view_html":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ","download_json":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ.json","view_paper":"https://pith.science/paper/SY5N2IVH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24192&json=true","fetch_graph":"https://pith.science/api/pith-number/SY5N2IVHSL2LHVA4JLUOBGVSWJ/graph.json","fetch_events":"https://pith.science/api/pith-number/SY5N2IVHSL2LHVA4JLUOBGVSWJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ/action/storage_attestation","attest_author":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ/action/author_attestation","sign_citation":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ/action/citation_signature","submit_replication":"https://pith.science/pith/SY5N2IVHSL2LHVA4JLUOBGVSWJ/action/replication_record"}},"created_at":"2026-05-26T01:02:51.706755+00:00","updated_at":"2026-05-26T01:02:51.706755+00:00"}