{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CX7D6XX5ZTUBOHKT7NC55ICKDQ","short_pith_number":"pith:CX7D6XX5","schema_version":"1.0","canonical_sha256":"15fe3f5efdcce8171d53fb45dea04a1c00dae42f130958990d3edc86b33f01f6","source":{"kind":"arxiv","id":"2407.15171","version":1},"attestation_state":"computed","paper":{"title":"Assessing Sample Quality via the Latent Space of Generative Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dimitris Samaras, Hieu Le, Jingyi Xu","submitted_at":"2024-07-21T14:05:06Z","abstract_excerpt":"Advances in generative models increase the need for sample quality assessment. To do so, previous methods rely on a pre-trained feature extractor to embed the generated samples and real samples into a common space for comparison. However, different feature extractors might lead to inconsistent assessment outcomes. Moreover, these methods are not applicable for domains where a robust, universal feature extractor does not yet exist, such as medical images or 3D assets. In this paper, we propose to directly examine the latent space of the trained generative model to infer generated sample quality"},"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":"2407.15171","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-07-21T14:05:06Z","cross_cats_sorted":[],"title_canon_sha256":"62023a7776641b76f2a0b6b8e610ac0148b9f9480ee53de058a5c01bbd0a306d","abstract_canon_sha256":"300090e2fbd669a20c61b15d52df9068fad2b513180d246fc0b643826f4624f6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:46:39.862073Z","signature_b64":"FYSX2J5F7UUq8PSTtE8+hEwookFHZ9n0mAQ+x5qQwM+EPS6Le3BhV+let9G6/QNIaVsqykGVCKHBNEOEjafLAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"15fe3f5efdcce8171d53fb45dea04a1c00dae42f130958990d3edc86b33f01f6","last_reissued_at":"2026-07-05T08:46:39.861606Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:46:39.861606Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Assessing Sample Quality via the Latent Space of Generative Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dimitris Samaras, Hieu Le, Jingyi Xu","submitted_at":"2024-07-21T14:05:06Z","abstract_excerpt":"Advances in generative models increase the need for sample quality assessment. To do so, previous methods rely on a pre-trained feature extractor to embed the generated samples and real samples into a common space for comparison. However, different feature extractors might lead to inconsistent assessment outcomes. Moreover, these methods are not applicable for domains where a robust, universal feature extractor does not yet exist, such as medical images or 3D assets. In this paper, we propose to directly examine the latent space of the trained generative model to infer generated sample quality"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.15171","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/2407.15171/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":"2407.15171","created_at":"2026-07-05T08:46:39.861669+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.15171v1","created_at":"2026-07-05T08:46:39.861669+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.15171","created_at":"2026-07-05T08:46:39.861669+00:00"},{"alias_kind":"pith_short_12","alias_value":"CX7D6XX5ZTUB","created_at":"2026-07-05T08:46:39.861669+00:00"},{"alias_kind":"pith_short_16","alias_value":"CX7D6XX5ZTUBOHKT","created_at":"2026-07-05T08:46:39.861669+00:00"},{"alias_kind":"pith_short_8","alias_value":"CX7D6XX5","created_at":"2026-07-05T08:46:39.861669+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.18810","citing_title":"All Seeds Are Not Equal: Enhancing Compositional Text-to-Image Generation with Reliable Random Seeds","ref_index":45,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ","json":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ.json","graph_json":"https://pith.science/api/pith-number/CX7D6XX5ZTUBOHKT7NC55ICKDQ/graph.json","events_json":"https://pith.science/api/pith-number/CX7D6XX5ZTUBOHKT7NC55ICKDQ/events.json","paper":"https://pith.science/paper/CX7D6XX5"},"agent_actions":{"view_html":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ","download_json":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ.json","view_paper":"https://pith.science/paper/CX7D6XX5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.15171&json=true","fetch_graph":"https://pith.science/api/pith-number/CX7D6XX5ZTUBOHKT7NC55ICKDQ/graph.json","fetch_events":"https://pith.science/api/pith-number/CX7D6XX5ZTUBOHKT7NC55ICKDQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ/action/storage_attestation","attest_author":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ/action/author_attestation","sign_citation":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ/action/citation_signature","submit_replication":"https://pith.science/pith/CX7D6XX5ZTUBOHKT7NC55ICKDQ/action/replication_record"}},"created_at":"2026-07-05T08:46:39.861669+00:00","updated_at":"2026-07-05T08:46:39.861669+00:00"}