{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:L436ILTOZ6NK3BLWCBVMU5R2LS","short_pith_number":"pith:L436ILTO","schema_version":"1.0","canonical_sha256":"5f37e42e6ecf9aad8576106aca763a5c81c32244c80acecf0e60c2f0dc06b8c9","source":{"kind":"arxiv","id":"2303.16887","version":2},"attestation_state":"computed","paper":{"title":"Towards Understanding the Effect of Pretraining Label Granularity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Ariel Fuxman, Enming Luo, Guan Zhe Hong, Stanley H. Chan, Yin Cui","submitted_at":"2023-03-29T17:56:36Z","abstract_excerpt":"In this paper, we study how the granularity of pretraining labels affects the generalization of deep neural networks in image classification tasks. We focus on the \"fine-to-coarse\" transfer learning setting, where the pretraining label space is more fine-grained than that of the target problem. Empirically, we show that pretraining on the leaf labels of ImageNet21k produces better transfer results on ImageNet1k than pretraining on other coarser granularity levels, which supports the common practice used in the community. Theoretically, we explain the benefit of fine-grained pretraining by prov"},"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":"2303.16887","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-03-29T17:56:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1a30486715ff5749eef0d26b75e286db24e3b6bd550074ef6fb9b7f827a7754f","abstract_canon_sha256":"5843712c854ae66b5afb8e8e559e12af005a7411b0679e30e360bd00ac9a7da0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:57:22.249111Z","signature_b64":"Ztud4oh1qx6VYjAlqQIWNGAf9c0w1ko9cdoq9VuKPliGICEXpxfTQSGk8r11XqNBOP2+idjPOQIV/R23+eJACQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f37e42e6ecf9aad8576106aca763a5c81c32244c80acecf0e60c2f0dc06b8c9","last_reissued_at":"2026-07-05T06:57:22.248739Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:57:22.248739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Understanding the Effect of Pretraining Label Granularity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Ariel Fuxman, Enming Luo, Guan Zhe Hong, Stanley H. Chan, Yin Cui","submitted_at":"2023-03-29T17:56:36Z","abstract_excerpt":"In this paper, we study how the granularity of pretraining labels affects the generalization of deep neural networks in image classification tasks. We focus on the \"fine-to-coarse\" transfer learning setting, where the pretraining label space is more fine-grained than that of the target problem. Empirically, we show that pretraining on the leaf labels of ImageNet21k produces better transfer results on ImageNet1k than pretraining on other coarser granularity levels, which supports the common practice used in the community. Theoretically, we explain the benefit of fine-grained pretraining by prov"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.16887","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/2303.16887/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":"2303.16887","created_at":"2026-07-05T06:57:22.248794+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.16887v2","created_at":"2026-07-05T06:57:22.248794+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.16887","created_at":"2026-07-05T06:57:22.248794+00:00"},{"alias_kind":"pith_short_12","alias_value":"L436ILTOZ6NK","created_at":"2026-07-05T06:57:22.248794+00:00"},{"alias_kind":"pith_short_16","alias_value":"L436ILTOZ6NK3BLW","created_at":"2026-07-05T06:57:22.248794+00:00"},{"alias_kind":"pith_short_8","alias_value":"L436ILTO","created_at":"2026-07-05T06:57:22.248794+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS","json":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS.json","graph_json":"https://pith.science/api/pith-number/L436ILTOZ6NK3BLWCBVMU5R2LS/graph.json","events_json":"https://pith.science/api/pith-number/L436ILTOZ6NK3BLWCBVMU5R2LS/events.json","paper":"https://pith.science/paper/L436ILTO"},"agent_actions":{"view_html":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS","download_json":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS.json","view_paper":"https://pith.science/paper/L436ILTO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.16887&json=true","fetch_graph":"https://pith.science/api/pith-number/L436ILTOZ6NK3BLWCBVMU5R2LS/graph.json","fetch_events":"https://pith.science/api/pith-number/L436ILTOZ6NK3BLWCBVMU5R2LS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS/action/storage_attestation","attest_author":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS/action/author_attestation","sign_citation":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS/action/citation_signature","submit_replication":"https://pith.science/pith/L436ILTOZ6NK3BLWCBVMU5R2LS/action/replication_record"}},"created_at":"2026-07-05T06:57:22.248794+00:00","updated_at":"2026-07-05T06:57:22.248794+00:00"}