{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:H5GEAT5Q47NTGSPSAA6JP5PPDE","short_pith_number":"pith:H5GEAT5Q","schema_version":"1.0","canonical_sha256":"3f4c404fb0e7db3349f2003c97f5ef19107dff86e5c77acbc37d46151587c346","source":{"kind":"arxiv","id":"2303.03032","version":1},"attestation_state":"computed","paper":{"title":"DeCap: Decoding CLIP Latents for Zero-Shot Captioning via Text-Only Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Linchao Zhu, Longyin Wen, Wei Li, Yi Yang","submitted_at":"2023-03-06T11:02:47Z","abstract_excerpt":"Large-scale pre-trained multi-modal models (e.g., CLIP) demonstrate strong zero-shot transfer capability in many discriminative tasks. Their adaptation to zero-shot image-conditioned text generation tasks has drawn increasing interest. Prior arts approach to zero-shot captioning by either utilizing the existing large language models (e.g., GPT-2) or pre-training the encoder-decoder network in an end-to-end manner. In this work, we propose a simple framework, named DeCap, for zero-shot captioning. We introduce a lightweight visual-aware language decoder. This decoder is both data-efficient and "},"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":"2303.03032","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-03-06T11:02:47Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"3c19aa379d366c7cab940f6b80fe2397febb3cb02ba1cf965c851b4db5fe4c8a","abstract_canon_sha256":"c65f61a2f07033e5842675815b9b645049a494e059f4d7de43bd0e69e1f8c2c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:48:18.462072Z","signature_b64":"d6QDkdp35rp1/vKKURT/C1o6QJ9KEfcphdp3ZYqXlVIxtnVD9QgivHCV44qKcKb4DzDSbTk1LPcx9YDlQA+tCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3f4c404fb0e7db3349f2003c97f5ef19107dff86e5c77acbc37d46151587c346","last_reissued_at":"2026-07-05T05:48:18.461693Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:48:18.461693Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeCap: Decoding CLIP Latents for Zero-Shot Captioning via Text-Only Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.CV","authors_text":"Linchao Zhu, Longyin Wen, Wei Li, Yi Yang","submitted_at":"2023-03-06T11:02:47Z","abstract_excerpt":"Large-scale pre-trained multi-modal models (e.g., CLIP) demonstrate strong zero-shot transfer capability in many discriminative tasks. Their adaptation to zero-shot image-conditioned text generation tasks has drawn increasing interest. Prior arts approach to zero-shot captioning by either utilizing the existing large language models (e.g., GPT-2) or pre-training the encoder-decoder network in an end-to-end manner. In this work, we propose a simple framework, named DeCap, for zero-shot captioning. We introduce a lightweight visual-aware language decoder. This decoder is both data-efficient and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.03032","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/2303.03032/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":"2303.03032","created_at":"2026-07-05T05:48:18.461747+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.03032v1","created_at":"2026-07-05T05:48:18.461747+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.03032","created_at":"2026-07-05T05:48:18.461747+00:00"},{"alias_kind":"pith_short_12","alias_value":"H5GEAT5Q47NT","created_at":"2026-07-05T05:48:18.461747+00:00"},{"alias_kind":"pith_short_16","alias_value":"H5GEAT5Q47NTGSPS","created_at":"2026-07-05T05:48:18.461747+00:00"},{"alias_kind":"pith_short_8","alias_value":"H5GEAT5Q","created_at":"2026-07-05T05:48:18.461747+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.07026","citing_title":"Modality Gap-Driven Subspace Alignment Training Paradigm For Multimodal Large Language Models","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE","json":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE.json","graph_json":"https://pith.science/api/pith-number/H5GEAT5Q47NTGSPSAA6JP5PPDE/graph.json","events_json":"https://pith.science/api/pith-number/H5GEAT5Q47NTGSPSAA6JP5PPDE/events.json","paper":"https://pith.science/paper/H5GEAT5Q"},"agent_actions":{"view_html":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE","download_json":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE.json","view_paper":"https://pith.science/paper/H5GEAT5Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.03032&json=true","fetch_graph":"https://pith.science/api/pith-number/H5GEAT5Q47NTGSPSAA6JP5PPDE/graph.json","fetch_events":"https://pith.science/api/pith-number/H5GEAT5Q47NTGSPSAA6JP5PPDE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE/action/storage_attestation","attest_author":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE/action/author_attestation","sign_citation":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE/action/citation_signature","submit_replication":"https://pith.science/pith/H5GEAT5Q47NTGSPSAA6JP5PPDE/action/replication_record"}},"created_at":"2026-07-05T05:48:18.461747+00:00","updated_at":"2026-07-05T05:48:18.461747+00:00"}