{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:F4DKGRLN5Y6B4ZSMD55YQ44P4U","short_pith_number":"pith:F4DKGRLN","schema_version":"1.0","canonical_sha256":"2f06a3456dee3c1e664c1f7b88738fe5300de641971ad67d64c3048caeaec30b","source":{"kind":"arxiv","id":"2409.02426","version":5},"attestation_state":"computed","paper":{"title":"Breaking the Curse of Dimensionality: Diffusion Models Efficiently Learn Low-Dimensional Distributions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Huijie Zhang, Peng Wang, Qing Qu, Siyi Chen, Yi Ma, Zekai Zhang","submitted_at":"2024-09-04T04:14:02Z","abstract_excerpt":"Despite their empirical success across a wide range of generative tasks, the fundamental principles underlying the ability of diffusion models to learn data distributions are poorly understood. In this work, we develop a new mathematical framework that explains how diffusion models can effectively learn low-dimensional distributions from a finite number of training samples without suffering from the curse of dimensionality. Specifically, motivated by the intrinsic low-dimensional structure of image data, we theoretically analyze a setting in which the data distribution is modeled as a mixture "},"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":"2409.02426","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-09-04T04:14:02Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"b57a38c5108cf4bc190afcc56da459bbe1b7cf4ecd433273dae39df99d83cb08","abstract_canon_sha256":"e25f5828cdb21cf4d1d87ccd6a588a83875e7e620b69b09871fe9bac915eb9f9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:09:12.894452Z","signature_b64":"xcA1d2JRVBqEGI2nwr8dg1cA4yBCcHXFnXoU93jrYQ8dQpn+APEgDsDVF1qiz27XZFpSYO1GIqEvu+KSBWXSDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f06a3456dee3c1e664c1f7b88738fe5300de641971ad67d64c3048caeaec30b","last_reissued_at":"2026-06-10T01:09:12.893413Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:09:12.893413Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Breaking the Curse of Dimensionality: Diffusion Models Efficiently Learn Low-Dimensional Distributions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Huijie Zhang, Peng Wang, Qing Qu, Siyi Chen, Yi Ma, Zekai Zhang","submitted_at":"2024-09-04T04:14:02Z","abstract_excerpt":"Despite their empirical success across a wide range of generative tasks, the fundamental principles underlying the ability of diffusion models to learn data distributions are poorly understood. In this work, we develop a new mathematical framework that explains how diffusion models can effectively learn low-dimensional distributions from a finite number of training samples without suffering from the curse of dimensionality. Specifically, motivated by the intrinsic low-dimensional structure of image data, we theoretically analyze a setting in which the data distribution is modeled as a mixture "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.02426","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/2409.02426/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":"2409.02426","created_at":"2026-06-10T01:09:12.893556+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.02426v5","created_at":"2026-06-10T01:09:12.893556+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.02426","created_at":"2026-06-10T01:09:12.893556+00:00"},{"alias_kind":"pith_short_12","alias_value":"F4DKGRLN5Y6B","created_at":"2026-06-10T01:09:12.893556+00:00"},{"alias_kind":"pith_short_16","alias_value":"F4DKGRLN5Y6B4ZSM","created_at":"2026-06-10T01:09:12.893556+00:00"},{"alias_kind":"pith_short_8","alias_value":"F4DKGRLN","created_at":"2026-06-10T01:09:12.893556+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2605.15822","citing_title":"Intrinsic Wasserstein Rates for Score-Based Generative Models on Smooth Manifolds","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06367","citing_title":"The Interplay of Data Structure and Imbalance in the Learning Dynamics of Diffusion Models","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10074","citing_title":"Transformers Learn the Optimal DDPM Denoiser for Multi-Token GMMs","ref_index":59,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U","json":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U.json","graph_json":"https://pith.science/api/pith-number/F4DKGRLN5Y6B4ZSMD55YQ44P4U/graph.json","events_json":"https://pith.science/api/pith-number/F4DKGRLN5Y6B4ZSMD55YQ44P4U/events.json","paper":"https://pith.science/paper/F4DKGRLN"},"agent_actions":{"view_html":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U","download_json":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U.json","view_paper":"https://pith.science/paper/F4DKGRLN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.02426&json=true","fetch_graph":"https://pith.science/api/pith-number/F4DKGRLN5Y6B4ZSMD55YQ44P4U/graph.json","fetch_events":"https://pith.science/api/pith-number/F4DKGRLN5Y6B4ZSMD55YQ44P4U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U/action/storage_attestation","attest_author":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U/action/author_attestation","sign_citation":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U/action/citation_signature","submit_replication":"https://pith.science/pith/F4DKGRLN5Y6B4ZSMD55YQ44P4U/action/replication_record"}},"created_at":"2026-06-10T01:09:12.893556+00:00","updated_at":"2026-06-10T01:09:12.893556+00:00"}