{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TV7XBSION7JMXII3DP463HYI2W","short_pith_number":"pith:TV7XBSIO","schema_version":"1.0","canonical_sha256":"9d7f70c90e6fd2cba11b1bf9ed9f08d5841fe1bbfe009e1b83eec5d5f3cd23c6","source":{"kind":"arxiv","id":"2206.01125","version":2},"attestation_state":"computed","paper":{"title":"Prefix Conditioning Unifies Language and Label Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen-Yu Lee, Chun-Liang Li, Kate Saenko, Kihyuk Sohn, Kuniaki Saito, Tomas Pfister, Xiang Zhang","submitted_at":"2022-06-02T16:12:26Z","abstract_excerpt":"Image-classification datasets have been used to pretrain image recognition models. Recently, web-scale image-caption datasets have emerged as a source of powerful pretraining alternative. Image-caption datasets are more ``open-domain'', containing a wider variety of scene types and vocabulary words than traditional classification datasets, and models trained on these datasets have demonstrated strong performance on few- and zero-shot recognition tasks. When naively unifying image-classification and -caption dataset, we show that such dataset biases negatively affect pre-training by reducing th"},"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":"2206.01125","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-06-02T16:12:26Z","cross_cats_sorted":[],"title_canon_sha256":"f588632538573771d774fb1326913a77f2a0e201cd2f09c72b2643adad385357","abstract_canon_sha256":"ad273139c4f389eb4c95742453e47733332bfa7d8e7e6380f9be05e26c482552"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:10:16.509042Z","signature_b64":"cZb4pqhwxUrdMLYLU03TqyC1M1hlw8kTmYxrOcs26EnLR8B+2s30I6ipDUjsQrGM8dJJ8aBeVE/3kAuKeZy1DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d7f70c90e6fd2cba11b1bf9ed9f08d5841fe1bbfe009e1b83eec5d5f3cd23c6","last_reissued_at":"2026-07-05T06:10:16.508645Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:10:16.508645Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prefix Conditioning Unifies Language and Label Supervision","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen-Yu Lee, Chun-Liang Li, Kate Saenko, Kihyuk Sohn, Kuniaki Saito, Tomas Pfister, Xiang Zhang","submitted_at":"2022-06-02T16:12:26Z","abstract_excerpt":"Image-classification datasets have been used to pretrain image recognition models. Recently, web-scale image-caption datasets have emerged as a source of powerful pretraining alternative. Image-caption datasets are more ``open-domain'', containing a wider variety of scene types and vocabulary words than traditional classification datasets, and models trained on these datasets have demonstrated strong performance on few- and zero-shot recognition tasks. When naively unifying image-classification and -caption dataset, we show that such dataset biases negatively affect pre-training by reducing th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.01125","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/2206.01125/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":"2206.01125","created_at":"2026-07-05T06:10:16.508702+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.01125v2","created_at":"2026-07-05T06:10:16.508702+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.01125","created_at":"2026-07-05T06:10:16.508702+00:00"},{"alias_kind":"pith_short_12","alias_value":"TV7XBSION7JM","created_at":"2026-07-05T06:10:16.508702+00:00"},{"alias_kind":"pith_short_16","alias_value":"TV7XBSION7JMXII3","created_at":"2026-07-05T06:10:16.508702+00:00"},{"alias_kind":"pith_short_8","alias_value":"TV7XBSIO","created_at":"2026-07-05T06:10:16.508702+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/TV7XBSION7JMXII3DP463HYI2W","json":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W.json","graph_json":"https://pith.science/api/pith-number/TV7XBSION7JMXII3DP463HYI2W/graph.json","events_json":"https://pith.science/api/pith-number/TV7XBSION7JMXII3DP463HYI2W/events.json","paper":"https://pith.science/paper/TV7XBSIO"},"agent_actions":{"view_html":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W","download_json":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W.json","view_paper":"https://pith.science/paper/TV7XBSIO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.01125&json=true","fetch_graph":"https://pith.science/api/pith-number/TV7XBSION7JMXII3DP463HYI2W/graph.json","fetch_events":"https://pith.science/api/pith-number/TV7XBSION7JMXII3DP463HYI2W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W/action/storage_attestation","attest_author":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W/action/author_attestation","sign_citation":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W/action/citation_signature","submit_replication":"https://pith.science/pith/TV7XBSION7JMXII3DP463HYI2W/action/replication_record"}},"created_at":"2026-07-05T06:10:16.508702+00:00","updated_at":"2026-07-05T06:10:16.508702+00:00"}