{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:H2VTEQBLUOR7EAWE22YFDXF6RE","short_pith_number":"pith:H2VTEQBL","schema_version":"1.0","canonical_sha256":"3eab32402ba3a3f202c4d6b051dcbe8913b4cb4bdb7e26e0c19ef4cf45f94f0b","source":{"kind":"arxiv","id":"2310.05473","version":1},"attestation_state":"computed","paper":{"title":"Sentence-level Prompts Benefit Composed Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chun-Mei Feng, Fahad Khan, Rick Siow Mong Goh, Salman Khan, Wangmeng Zuo, Xinxing Xu, Yang Bai, Yong Liu","submitted_at":"2023-10-09T07:31:44Z","abstract_excerpt":"Composed image retrieval (CIR) is the task of retrieving specific images by using a query that involves both a reference image and a relative caption. Most existing CIR models adopt the late-fusion strategy to combine visual and language features. Besides, several approaches have also been suggested to generate a pseudo-word token from the reference image, which is further integrated into the relative caption for CIR. However, these pseudo-word-based prompting methods have limitations when target image encompasses complex changes on reference image, e.g., object removal and attribute modificat"},"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":"2310.05473","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-10-09T07:31:44Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"81d2bae85b32e95a41c849bcafa8f4dada1818b6da0b67e5fbb5fb49020323ad","abstract_canon_sha256":"3cc6f8768bc1b3ce0bc3fe4a1bedc94aa01b154d5a9a881341d32d9e56895d85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:58:45.777510Z","signature_b64":"hQrO6HweJrMb0y2RCoV01ubrpdMffXHvLhTdZ9yUaWM566ITWonk3B2CHzxCJEPKbDruLhE5meZyEYbkxbb8Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3eab32402ba3a3f202c4d6b051dcbe8913b4cb4bdb7e26e0c19ef4cf45f94f0b","last_reissued_at":"2026-07-05T06:58:45.776923Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:58:45.776923Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sentence-level Prompts Benefit Composed Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chun-Mei Feng, Fahad Khan, Rick Siow Mong Goh, Salman Khan, Wangmeng Zuo, Xinxing Xu, Yang Bai, Yong Liu","submitted_at":"2023-10-09T07:31:44Z","abstract_excerpt":"Composed image retrieval (CIR) is the task of retrieving specific images by using a query that involves both a reference image and a relative caption. Most existing CIR models adopt the late-fusion strategy to combine visual and language features. Besides, several approaches have also been suggested to generate a pseudo-word token from the reference image, which is further integrated into the relative caption for CIR. However, these pseudo-word-based prompting methods have limitations when target image encompasses complex changes on reference image, e.g., object removal and attribute modificat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.05473","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/2310.05473/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":"2310.05473","created_at":"2026-07-05T06:58:45.776992+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.05473v1","created_at":"2026-07-05T06:58:45.776992+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.05473","created_at":"2026-07-05T06:58:45.776992+00:00"},{"alias_kind":"pith_short_12","alias_value":"H2VTEQBLUOR7","created_at":"2026-07-05T06:58:45.776992+00:00"},{"alias_kind":"pith_short_16","alias_value":"H2VTEQBLUOR7EAWE","created_at":"2026-07-05T06:58:45.776992+00:00"},{"alias_kind":"pith_short_8","alias_value":"H2VTEQBL","created_at":"2026-07-05T06:58:45.776992+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19684","citing_title":"Exploring Multi-Modal Large Language Models and Two-Stage Fine-Tuning for Fashion Image Retrieval","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08144","citing_title":"IMAGINE: Adaptive Schema-Imagery Enhanced Composition for Composed Video Retrieval","ref_index":109,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05809","citing_title":"Stealthy and Adjustable Text-Guided Backdoor Attacks on Multimodal Pretrained Models","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE","json":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE.json","graph_json":"https://pith.science/api/pith-number/H2VTEQBLUOR7EAWE22YFDXF6RE/graph.json","events_json":"https://pith.science/api/pith-number/H2VTEQBLUOR7EAWE22YFDXF6RE/events.json","paper":"https://pith.science/paper/H2VTEQBL"},"agent_actions":{"view_html":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE","download_json":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE.json","view_paper":"https://pith.science/paper/H2VTEQBL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.05473&json=true","fetch_graph":"https://pith.science/api/pith-number/H2VTEQBLUOR7EAWE22YFDXF6RE/graph.json","fetch_events":"https://pith.science/api/pith-number/H2VTEQBLUOR7EAWE22YFDXF6RE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE/action/storage_attestation","attest_author":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE/action/author_attestation","sign_citation":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE/action/citation_signature","submit_replication":"https://pith.science/pith/H2VTEQBLUOR7EAWE22YFDXF6RE/action/replication_record"}},"created_at":"2026-07-05T06:58:45.776992+00:00","updated_at":"2026-07-05T06:58:45.776992+00:00"}