{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:C4WCZ3FFO6MPS4GDJFWPEUX63L","short_pith_number":"pith:C4WCZ3FF","schema_version":"1.0","canonical_sha256":"172c2ceca57798f970c3496cf252fedaf7010718e111be836e30701a627f6332","source":{"kind":"arxiv","id":"2311.00056","version":1},"attestation_state":"computed","paper":{"title":"Diversity and Diffusion: Observations on Synthetic Image Distributions with Stable Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"David Marwood, Shumeet Baluja, Yair Alon","submitted_at":"2023-10-31T18:05:15Z","abstract_excerpt":"Recent progress in text-to-image (TTI) systems, such as StableDiffusion, Imagen, and DALL-E 2, have made it possible to create realistic images with simple text prompts. It is tempting to use these systems to eliminate the manual task of obtaining natural images for training a new machine learning classifier. However, in all of the experiments performed to date, classifiers trained solely with synthetic images perform poorly at inference, despite the images used for training appearing realistic. Examining this apparent incongruity in detail gives insight into the limitations of the underlying "},"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":"2311.00056","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-10-31T18:05:15Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"8f486d7181e907f29fb7288ad029e9640d6fcda36ddca489d5ae224924fd68f0","abstract_canon_sha256":"7ccce41a54f81d4b338a73af03602639973494c7e2b55488e8f6b086ce619eb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:07:41.543331Z","signature_b64":"HBJSIoSaOJvC+BRLldcpKA+ulZ+GK+lev8QquS7nDyC2I2lSHJTNPw+fIdo0anJ85VIfDe8sSpBTiQxKSqV3Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"172c2ceca57798f970c3496cf252fedaf7010718e111be836e30701a627f6332","last_reissued_at":"2026-07-05T07:07:41.542860Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:07:41.542860Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diversity and Diffusion: Observations on Synthetic Image Distributions with Stable Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"David Marwood, Shumeet Baluja, Yair Alon","submitted_at":"2023-10-31T18:05:15Z","abstract_excerpt":"Recent progress in text-to-image (TTI) systems, such as StableDiffusion, Imagen, and DALL-E 2, have made it possible to create realistic images with simple text prompts. It is tempting to use these systems to eliminate the manual task of obtaining natural images for training a new machine learning classifier. However, in all of the experiments performed to date, classifiers trained solely with synthetic images perform poorly at inference, despite the images used for training appearing realistic. Examining this apparent incongruity in detail gives insight into the limitations of the underlying "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.00056","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/2311.00056/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":"2311.00056","created_at":"2026-07-05T07:07:41.542914+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.00056v1","created_at":"2026-07-05T07:07:41.542914+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.00056","created_at":"2026-07-05T07:07:41.542914+00:00"},{"alias_kind":"pith_short_12","alias_value":"C4WCZ3FFO6MP","created_at":"2026-07-05T07:07:41.542914+00:00"},{"alias_kind":"pith_short_16","alias_value":"C4WCZ3FFO6MPS4GD","created_at":"2026-07-05T07:07:41.542914+00:00"},{"alias_kind":"pith_short_8","alias_value":"C4WCZ3FF","created_at":"2026-07-05T07:07:41.542914+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.06248","citing_title":"Rendering-Refined Stable Diffusion for Privacy Compliant Synthetic Data","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L","json":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L.json","graph_json":"https://pith.science/api/pith-number/C4WCZ3FFO6MPS4GDJFWPEUX63L/graph.json","events_json":"https://pith.science/api/pith-number/C4WCZ3FFO6MPS4GDJFWPEUX63L/events.json","paper":"https://pith.science/paper/C4WCZ3FF"},"agent_actions":{"view_html":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L","download_json":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L.json","view_paper":"https://pith.science/paper/C4WCZ3FF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.00056&json=true","fetch_graph":"https://pith.science/api/pith-number/C4WCZ3FFO6MPS4GDJFWPEUX63L/graph.json","fetch_events":"https://pith.science/api/pith-number/C4WCZ3FFO6MPS4GDJFWPEUX63L/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L/action/storage_attestation","attest_author":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L/action/author_attestation","sign_citation":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L/action/citation_signature","submit_replication":"https://pith.science/pith/C4WCZ3FFO6MPS4GDJFWPEUX63L/action/replication_record"}},"created_at":"2026-07-05T07:07:41.542914+00:00","updated_at":"2026-07-05T07:07:41.542914+00:00"}