{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7MEWYLCWL4JW5FFIMPGX7PVICW","short_pith_number":"pith:7MEWYLCW","schema_version":"1.0","canonical_sha256":"fb096c2c565f136e94a863cd7fbea815bae49a36dc9f4e75576e36245a9a0751","source":{"kind":"arxiv","id":"2502.20946","version":2},"attestation_state":"computed","paper":{"title":"Generative Uncertainty in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dan Zhang, Eliot Wong-Toi, Eric Nalisnick, Guoxuan Xia, Metod Jazbec, Stephan Mandt","submitted_at":"2025-02-28T10:56:39Z","abstract_excerpt":"Diffusion models have recently driven significant breakthroughs in generative modeling. While state-of-the-art models produce high-quality samples on average, individual samples can still be low quality. Detecting such samples without human inspection remains a challenging task. To address this, we propose a Bayesian framework for estimating generative uncertainty of synthetic samples. We outline how to make Bayesian inference practical for large, modern generative models and introduce a new semantic likelihood (evaluated in the latent space of a feature extractor) to address the challenges po"},"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":"2502.20946","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-28T10:56:39Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"97aeec2651f48f25b95b2eec466e9400d9f0f3f2b7810ce83950fc77c52db391","abstract_canon_sha256":"2ce23a4e242ea4abb0c59305c56dd9d122c0e0499f25c569139cc7a86730d9ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:10.346103Z","signature_b64":"NjSXFbcZZNXZMsN59L4kUCyZhwFyOnjKN8O6R2LOZuQhSy4oCA3vUZONO6ovgXuJAcAbD2lSPn03M87ZlXE2DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb096c2c565f136e94a863cd7fbea815bae49a36dc9f4e75576e36245a9a0751","last_reissued_at":"2026-07-05T11:20:10.345643Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:10.345643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generative Uncertainty in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dan Zhang, Eliot Wong-Toi, Eric Nalisnick, Guoxuan Xia, Metod Jazbec, Stephan Mandt","submitted_at":"2025-02-28T10:56:39Z","abstract_excerpt":"Diffusion models have recently driven significant breakthroughs in generative modeling. While state-of-the-art models produce high-quality samples on average, individual samples can still be low quality. Detecting such samples without human inspection remains a challenging task. To address this, we propose a Bayesian framework for estimating generative uncertainty of synthetic samples. We outline how to make Bayesian inference practical for large, modern generative models and introduce a new semantic likelihood (evaluated in the latent space of a feature extractor) to address the challenges po"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20946","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/2502.20946/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":"2502.20946","created_at":"2026-07-05T11:20:10.345701+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20946v2","created_at":"2026-07-05T11:20:10.345701+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20946","created_at":"2026-07-05T11:20:10.345701+00:00"},{"alias_kind":"pith_short_12","alias_value":"7MEWYLCWL4JW","created_at":"2026-07-05T11:20:10.345701+00:00"},{"alias_kind":"pith_short_16","alias_value":"7MEWYLCWL4JW5FFI","created_at":"2026-07-05T11:20:10.345701+00:00"},{"alias_kind":"pith_short_8","alias_value":"7MEWYLCW","created_at":"2026-07-05T11:20:10.345701+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01595","citing_title":"Uncertainty-Calibrated Diffusion for Reliable 3D Molecular Graph Generation","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00941","citing_title":"Divergence is Uncertainty: A Closed-Form Posterior Covariance for Flow Matching","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2603.21717","citing_title":"Uncertainty-Aware Distribution-to-Distribution Flow Matching for Scientific Imaging","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00941","citing_title":"Divergence is Uncertainty: A Closed-Form Posterior Covariance for Flow Matching","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18472","citing_title":"Flowing with Confidence","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2603.21717","citing_title":"Uncertainty-Aware Distribution-to-Distribution Flow Matching for Scientific Imaging","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09031","citing_title":"Spherical Boltzmann machines: a solvable theory of learning and generation in energy-based models","ref_index":113,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17936","citing_title":"Replica Theory of Spherical Boltzmann Machine Ensembles","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW","json":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW.json","graph_json":"https://pith.science/api/pith-number/7MEWYLCWL4JW5FFIMPGX7PVICW/graph.json","events_json":"https://pith.science/api/pith-number/7MEWYLCWL4JW5FFIMPGX7PVICW/events.json","paper":"https://pith.science/paper/7MEWYLCW"},"agent_actions":{"view_html":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW","download_json":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW.json","view_paper":"https://pith.science/paper/7MEWYLCW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20946&json=true","fetch_graph":"https://pith.science/api/pith-number/7MEWYLCWL4JW5FFIMPGX7PVICW/graph.json","fetch_events":"https://pith.science/api/pith-number/7MEWYLCWL4JW5FFIMPGX7PVICW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW/action/storage_attestation","attest_author":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW/action/author_attestation","sign_citation":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW/action/citation_signature","submit_replication":"https://pith.science/pith/7MEWYLCWL4JW5FFIMPGX7PVICW/action/replication_record"}},"created_at":"2026-07-05T11:20:10.345701+00:00","updated_at":"2026-07-05T11:20:10.345701+00:00"}