{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KT6DH7YVCZ4ZQVUP3ZFMZPAE63","short_pith_number":"pith:KT6DH7YV","schema_version":"1.0","canonical_sha256":"54fc33ff15167998568fde4accbc04f6d28da4521a8cdf11f91ac7d5f85dcde7","source":{"kind":"arxiv","id":"2502.02483","version":3},"attestation_state":"computed","paper":{"title":"Distributional Diffusion Models with Scoring Rules","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexandre Galashov, Arnaud Doucet, Arthur Gretton, Guangyao Zhou, J. Swaroop Guntupalli, Kevin Murphy, Valentin De Bortoli","submitted_at":"2025-02-04T16:59:03Z","abstract_excerpt":"Diffusion models generate high-quality synthetic data. They operate by defining a continuous-time forward process which gradually adds Gaussian noise to data until fully corrupted. The corresponding reverse process progressively \"denoises\" a Gaussian sample into a sample from the data distribution. However, generating high-quality outputs requires many discretization steps to obtain a faithful approximation of the reverse process. This is expensive and has motivated the development of many acceleration methods. We propose to accomplish sample generation by learning the posterior {\\em distribut"},"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":"2502.02483","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-04T16:59:03Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"611d80926fb0ec210ff3aa9d034a3ae06cbf1ec6b7fda7c37b48b6a4bb07287d","abstract_canon_sha256":"adeb911dde0d7deded8b3683d810f679e52ae234ae47a78877372e8f09dddeb9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:03.631394Z","signature_b64":"GMZbT0Jw9M+MqkuRU5iYjK/KNVfH4uRp6dADs77cDXsPSU/WmkE8rHYtsXSoyyfplh6CvSzxxaWdv2dhgT2FBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"54fc33ff15167998568fde4accbc04f6d28da4521a8cdf11f91ac7d5f85dcde7","last_reissued_at":"2026-07-05T11:33:03.630888Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:03.630888Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distributional Diffusion Models with Scoring Rules","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexandre Galashov, Arnaud Doucet, Arthur Gretton, Guangyao Zhou, J. Swaroop Guntupalli, Kevin Murphy, Valentin De Bortoli","submitted_at":"2025-02-04T16:59:03Z","abstract_excerpt":"Diffusion models generate high-quality synthetic data. They operate by defining a continuous-time forward process which gradually adds Gaussian noise to data until fully corrupted. The corresponding reverse process progressively \"denoises\" a Gaussian sample into a sample from the data distribution. However, generating high-quality outputs requires many discretization steps to obtain a faithful approximation of the reverse process. This is expensive and has motivated the development of many acceleration methods. We propose to accomplish sample generation by learning the posterior {\\em distribut"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.02483","kind":"arxiv","version":3},"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/2502.02483/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":"2502.02483","created_at":"2026-07-05T11:33:03.630950+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.02483v3","created_at":"2026-07-05T11:33:03.630950+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.02483","created_at":"2026-07-05T11:33:03.630950+00:00"},{"alias_kind":"pith_short_12","alias_value":"KT6DH7YVCZ4Z","created_at":"2026-07-05T11:33:03.630950+00:00"},{"alias_kind":"pith_short_16","alias_value":"KT6DH7YVCZ4ZQVUP","created_at":"2026-07-05T11:33:03.630950+00:00"},{"alias_kind":"pith_short_8","alias_value":"KT6DH7YV","created_at":"2026-07-05T11:33:03.630950+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10089","citing_title":"A Theory on Flow Matching with Neural Networks","ref_index":287,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01171","citing_title":"Decision-Aware Training for Sample-Based Generative Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06624","citing_title":"Principles and Practice of Deep Representation Learning: or a Mathematical Theory of Memory","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01002","citing_title":"Theoretical Analysis of Engression and Reverse Markov Engression","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63","json":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63.json","graph_json":"https://pith.science/api/pith-number/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/graph.json","events_json":"https://pith.science/api/pith-number/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/events.json","paper":"https://pith.science/paper/KT6DH7YV"},"agent_actions":{"view_html":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63","download_json":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63.json","view_paper":"https://pith.science/paper/KT6DH7YV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.02483&json=true","fetch_graph":"https://pith.science/api/pith-number/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/graph.json","fetch_events":"https://pith.science/api/pith-number/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/action/storage_attestation","attest_author":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/action/author_attestation","sign_citation":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/action/citation_signature","submit_replication":"https://pith.science/pith/KT6DH7YVCZ4ZQVUP3ZFMZPAE63/action/replication_record"}},"created_at":"2026-07-05T11:33:03.630950+00:00","updated_at":"2026-07-05T11:33:03.630950+00:00"}