{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IJUFY3ZG3H4MGAHOI3JDDPTH7W","short_pith_number":"pith:IJUFY3ZG","schema_version":"1.0","canonical_sha256":"42685c6f26d9f8c300ee46d231be67fd895a388ae089af86da239a644a125215","source":{"kind":"arxiv","id":"2404.18444","version":2},"attestation_state":"computed","paper":{"title":"U-Nets as Belief Propagation: Efficient Classification, Denoising, and Diffusion in Generative Hierarchical Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","math.ST","stat.ML","stat.TH"],"primary_cat":"cs.LG","authors_text":"Song Mei","submitted_at":"2024-04-29T05:57:03Z","abstract_excerpt":"U-Nets are among the most widely used architectures in computer vision, renowned for their exceptional performance in applications such as image segmentation, denoising, and diffusion modeling. However, a theoretical explanation of the U-Net architecture design has not yet been fully established.\n  This paper introduces a novel interpretation of the U-Net architecture by studying certain generative hierarchical models, which are tree-structured graphical models extensively utilized in both language and image domains. With their encoder-decoder structure, long skip connections, and pooling and "},"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":"2404.18444","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2024-04-29T05:57:03Z","cross_cats_sorted":["cs.AI","math.ST","stat.ML","stat.TH"],"title_canon_sha256":"0a3f97fa40f01460342d123510de4056b918ccb956d5e2f00bd741ffdcc3d7e9","abstract_canon_sha256":"0943f490ffd1be0f477a288d8620bb97293607ab9058392d002b54ab040467a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:14:06.390541Z","signature_b64":"ngppErbxMMMaGZ74O4krpN2nC5UcqND8/J7A1uSsKcVGPpEH3yEpiIOQfu8sSdWqpZ0aAU/BNLq/rRqKlSk1Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"42685c6f26d9f8c300ee46d231be67fd895a388ae089af86da239a644a125215","last_reissued_at":"2026-07-05T08:14:06.389994Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:14:06.389994Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"U-Nets as Belief Propagation: Efficient Classification, Denoising, and Diffusion in Generative Hierarchical Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","math.ST","stat.ML","stat.TH"],"primary_cat":"cs.LG","authors_text":"Song Mei","submitted_at":"2024-04-29T05:57:03Z","abstract_excerpt":"U-Nets are among the most widely used architectures in computer vision, renowned for their exceptional performance in applications such as image segmentation, denoising, and diffusion modeling. However, a theoretical explanation of the U-Net architecture design has not yet been fully established.\n  This paper introduces a novel interpretation of the U-Net architecture by studying certain generative hierarchical models, which are tree-structured graphical models extensively utilized in both language and image domains. With their encoder-decoder structure, long skip connections, and pooling and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.18444","kind":"arxiv","version":2},"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/2404.18444/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":"2404.18444","created_at":"2026-07-05T08:14:06.390051+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.18444v2","created_at":"2026-07-05T08:14:06.390051+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.18444","created_at":"2026-07-05T08:14:06.390051+00:00"},{"alias_kind":"pith_short_12","alias_value":"IJUFY3ZG3H4M","created_at":"2026-07-05T08:14:06.390051+00:00"},{"alias_kind":"pith_short_16","alias_value":"IJUFY3ZG3H4MGAHO","created_at":"2026-07-05T08:14:06.390051+00:00"},{"alias_kind":"pith_short_8","alias_value":"IJUFY3ZG","created_at":"2026-07-05T08:14:06.390051+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27734","citing_title":"Learn from your own latents and not from tokens: A sample-complexity theory","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2512.09806","citing_title":"CHEM: Estimating and Understanding Hallucinations in Deep Learning for Image Processing","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W","json":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W.json","graph_json":"https://pith.science/api/pith-number/IJUFY3ZG3H4MGAHOI3JDDPTH7W/graph.json","events_json":"https://pith.science/api/pith-number/IJUFY3ZG3H4MGAHOI3JDDPTH7W/events.json","paper":"https://pith.science/paper/IJUFY3ZG"},"agent_actions":{"view_html":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W","download_json":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W.json","view_paper":"https://pith.science/paper/IJUFY3ZG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.18444&json=true","fetch_graph":"https://pith.science/api/pith-number/IJUFY3ZG3H4MGAHOI3JDDPTH7W/graph.json","fetch_events":"https://pith.science/api/pith-number/IJUFY3ZG3H4MGAHOI3JDDPTH7W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W/action/storage_attestation","attest_author":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W/action/author_attestation","sign_citation":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W/action/citation_signature","submit_replication":"https://pith.science/pith/IJUFY3ZG3H4MGAHOI3JDDPTH7W/action/replication_record"}},"created_at":"2026-07-05T08:14:06.390051+00:00","updated_at":"2026-07-05T08:14:06.390051+00:00"}