{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:DXUCWEUDCD6X5HV2GBKIAXCS6F","short_pith_number":"pith:DXUCWEUD","schema_version":"1.0","canonical_sha256":"1de82b128310fd7e9eba3054805c52f142cd6c53d8c9e087c12ab12ce94b12b1","source":{"kind":"arxiv","id":"2307.07397","version":1},"attestation_state":"computed","paper":{"title":"Improving Zero-Shot Generalization for CLIP with Synthesized Prompts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jian Liang, Nan Xu, Ran He, Tieniu Tan, Zhengbo Wang, Zilei Wang","submitted_at":"2023-07-14T15:15:45Z","abstract_excerpt":"With the growing interest in pretrained vision-language models like CLIP, recent research has focused on adapting these models to downstream tasks. Despite achieving promising results, most existing methods require labeled data for all classes, which may not hold in real-world applications due to the long tail and Zipf's law. For example, some classes may lack labeled data entirely, such as emerging concepts. To address this problem, we propose a plug-and-play generative approach called \\textbf{S}ynt\\textbf{H}es\\textbf{I}zed \\textbf{P}rompts~(\\textbf{SHIP}) to improve existing fine-tuning meth"},"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":"2307.07397","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-07-14T15:15:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4116d2523da05df57a5fa4727addfd2f85d920a03cbf2c582305ac2eb8fc0232","abstract_canon_sha256":"5a8cb43afa0a3f60374929b73f033d5767af7dc9cccbd0421665f0f7412d7b8b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:30:55.088082Z","signature_b64":"lHe8cF03ZID6yHO4qYb5Lz9PchBdd0Gx1pBS7jonVXTfKRmyNJYucalYE+mmodT38GCFtwDAh4sgM1QX5PKrAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1de82b128310fd7e9eba3054805c52f142cd6c53d8c9e087c12ab12ce94b12b1","last_reissued_at":"2026-07-05T06:30:55.087594Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:30:55.087594Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Zero-Shot Generalization for CLIP with Synthesized Prompts","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jian Liang, Nan Xu, Ran He, Tieniu Tan, Zhengbo Wang, Zilei Wang","submitted_at":"2023-07-14T15:15:45Z","abstract_excerpt":"With the growing interest in pretrained vision-language models like CLIP, recent research has focused on adapting these models to downstream tasks. Despite achieving promising results, most existing methods require labeled data for all classes, which may not hold in real-world applications due to the long tail and Zipf's law. For example, some classes may lack labeled data entirely, such as emerging concepts. To address this problem, we propose a plug-and-play generative approach called \\textbf{S}ynt\\textbf{H}es\\textbf{I}zed \\textbf{P}rompts~(\\textbf{SHIP}) to improve existing fine-tuning meth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.07397","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/2307.07397/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":"2307.07397","created_at":"2026-07-05T06:30:55.087659+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.07397v1","created_at":"2026-07-05T06:30:55.087659+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.07397","created_at":"2026-07-05T06:30:55.087659+00:00"},{"alias_kind":"pith_short_12","alias_value":"DXUCWEUDCD6X","created_at":"2026-07-05T06:30:55.087659+00:00"},{"alias_kind":"pith_short_16","alias_value":"DXUCWEUDCD6X5HV2","created_at":"2026-07-05T06:30:55.087659+00:00"},{"alias_kind":"pith_short_8","alias_value":"DXUCWEUD","created_at":"2026-07-05T06:30:55.087659+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03730","citing_title":"Beyond False Stability: High-Noise Drift Gating for Test-Time Adversarial Defenses in Vision-Language Models","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25922","citing_title":"Closed-Loop Bidirectional Prompting for Adversarial Robustness of Vision Language Models","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2603.25383","citing_title":"CLIP-RD: Relative Distillation for Efficient CLIP Knowledge Distillation","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F","json":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F.json","graph_json":"https://pith.science/api/pith-number/DXUCWEUDCD6X5HV2GBKIAXCS6F/graph.json","events_json":"https://pith.science/api/pith-number/DXUCWEUDCD6X5HV2GBKIAXCS6F/events.json","paper":"https://pith.science/paper/DXUCWEUD"},"agent_actions":{"view_html":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F","download_json":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F.json","view_paper":"https://pith.science/paper/DXUCWEUD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.07397&json=true","fetch_graph":"https://pith.science/api/pith-number/DXUCWEUDCD6X5HV2GBKIAXCS6F/graph.json","fetch_events":"https://pith.science/api/pith-number/DXUCWEUDCD6X5HV2GBKIAXCS6F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F/action/storage_attestation","attest_author":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F/action/author_attestation","sign_citation":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F/action/citation_signature","submit_replication":"https://pith.science/pith/DXUCWEUDCD6X5HV2GBKIAXCS6F/action/replication_record"}},"created_at":"2026-07-05T06:30:55.087659+00:00","updated_at":"2026-07-05T06:30:55.087659+00:00"}