{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:UEFV3TYNIWFZLCYTTMFK5PKHDL","short_pith_number":"pith:UEFV3TYN","schema_version":"1.0","canonical_sha256":"a10b5dcf0d458b958b139b0aaebd471af8df78addb93d2cabf140098274eeae1","source":{"kind":"arxiv","id":"2401.09603","version":2},"attestation_state":"computed","paper":{"title":"Rethinking FID: Towards a Better Evaluation Metric for Image Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Andreas Veit, Ayan Chakrabarti, Daniel Glasner, Sadeep Jayasumana, Sanjiv Kumar, Srikumar Ramalingam","submitted_at":"2023-11-30T19:11:01Z","abstract_excerpt":"As with many machine learning problems, the progress of image generation methods hinges on good evaluation metrics. One of the most popular is the Frechet Inception Distance (FID). FID estimates the distance between a distribution of Inception-v3 features of real images, and those of images generated by the algorithm. We highlight important drawbacks of FID: Inception's poor representation of the rich and varied content generated by modern text-to-image models, incorrect normality assumptions, and poor sample complexity. We call for a reevaluation of FID's use as the primary quality metric for"},"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":"2401.09603","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2023-11-30T19:11:01Z","cross_cats_sorted":[],"title_canon_sha256":"a29a594bb58c2a9ee057e2ac60a165edb7e1843465689c6f9ddf208738ac72cc","abstract_canon_sha256":"d3854f4d9637c7f544853c7fe2dcf1df8dd176b01cc69634c343235aaa5de90f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:37:55.249083Z","signature_b64":"eiUb8ZsuicPQS3WiNkJKLSw4ZiFwsd66nfHHXfJp0IrHJYEo7dy2Up685nezZV9CFaeN7dUtl/7qVMohIpvpDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a10b5dcf0d458b958b139b0aaebd471af8df78addb93d2cabf140098274eeae1","last_reissued_at":"2026-07-05T07:37:55.248600Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:37:55.248600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking FID: Towards a Better Evaluation Metric for Image Generation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Andreas Veit, Ayan Chakrabarti, Daniel Glasner, Sadeep Jayasumana, Sanjiv Kumar, Srikumar Ramalingam","submitted_at":"2023-11-30T19:11:01Z","abstract_excerpt":"As with many machine learning problems, the progress of image generation methods hinges on good evaluation metrics. One of the most popular is the Frechet Inception Distance (FID). FID estimates the distance between a distribution of Inception-v3 features of real images, and those of images generated by the algorithm. We highlight important drawbacks of FID: Inception's poor representation of the rich and varied content generated by modern text-to-image models, incorrect normality assumptions, and poor sample complexity. We call for a reevaluation of FID's use as the primary quality metric for"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.09603","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/2401.09603/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":"2401.09603","created_at":"2026-07-05T07:37:55.248659+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.09603v2","created_at":"2026-07-05T07:37:55.248659+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.09603","created_at":"2026-07-05T07:37:55.248659+00:00"},{"alias_kind":"pith_short_12","alias_value":"UEFV3TYNIWFZ","created_at":"2026-07-05T07:37:55.248659+00:00"},{"alias_kind":"pith_short_16","alias_value":"UEFV3TYNIWFZLCYT","created_at":"2026-07-05T07:37:55.248659+00:00"},{"alias_kind":"pith_short_8","alias_value":"UEFV3TYN","created_at":"2026-07-05T07:37:55.248659+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12858","citing_title":"JSCGC: Joint Source-Channel-Generation Coding for Wireless Generative Communications","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29059","citing_title":"Flow Matching in Feature Space for Stochastic World Modeling","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02406","citing_title":"Evaluating AI-Generated Images of Cultural Artifacts with Community-Informed Rubrics","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2503.07703","citing_title":"Seedream 2.0: A Native Chinese-English Bilingual Image Generation Foundation Model","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02406","citing_title":"Evaluating AI-Generated Images of Cultural Artifacts with Community-Informed Rubrics","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05627","citing_title":"Leveraging Image Generators to Address Data Scarcity: The Gen4Regen Dataset for Forest Regeneration Mapping","ref_index":75,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL","json":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL.json","graph_json":"https://pith.science/api/pith-number/UEFV3TYNIWFZLCYTTMFK5PKHDL/graph.json","events_json":"https://pith.science/api/pith-number/UEFV3TYNIWFZLCYTTMFK5PKHDL/events.json","paper":"https://pith.science/paper/UEFV3TYN"},"agent_actions":{"view_html":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL","download_json":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL.json","view_paper":"https://pith.science/paper/UEFV3TYN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.09603&json=true","fetch_graph":"https://pith.science/api/pith-number/UEFV3TYNIWFZLCYTTMFK5PKHDL/graph.json","fetch_events":"https://pith.science/api/pith-number/UEFV3TYNIWFZLCYTTMFK5PKHDL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL/action/storage_attestation","attest_author":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL/action/author_attestation","sign_citation":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL/action/citation_signature","submit_replication":"https://pith.science/pith/UEFV3TYNIWFZLCYTTMFK5PKHDL/action/replication_record"}},"created_at":"2026-07-05T07:37:55.248659+00:00","updated_at":"2026-07-05T07:37:55.248659+00:00"}