{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:SVGDNFZ5JDPCXBNYBJY7TRMYE6","short_pith_number":"pith:SVGDNFZ5","schema_version":"1.0","canonical_sha256":"954c36973d48de2b85b80a71f9c59827b6386b59b7564d307d951de52b28f11b","source":{"kind":"arxiv","id":"2303.15247","version":2},"attestation_state":"computed","paper":{"title":"Zero-Shot Composed Image Retrieval with Textual Inversion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.IR"],"primary_cat":"cs.CV","authors_text":"Alberto Baldrati, Alberto Del Bimbo, Lorenzo Agnolucci, Marco Bertini","submitted_at":"2023-03-27T14:31:25Z","abstract_excerpt":"Composed Image Retrieval (CIR) aims to retrieve a target image based on a query composed of a reference image and a relative caption that describes the difference between the two images. The high effort and cost required for labeling datasets for CIR hamper the widespread usage of existing methods, as they rely on supervised learning. In this work, we propose a new task, Zero-Shot CIR (ZS-CIR), that aims to address CIR without requiring a labeled training dataset. Our approach, named zero-Shot composEd imAge Retrieval with textuaL invErsion (SEARLE), maps the visual features of the reference i"},"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":"2303.15247","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-03-27T14:31:25Z","cross_cats_sorted":["cs.CL","cs.IR"],"title_canon_sha256":"1d799c2c178477a39ebc89ef0e5a5efd8c53be9e78ed8baed0f0a60fed9391fc","abstract_canon_sha256":"c51831870f40ab977da0fa0989c8ced8ac30f4a569e970e157e2bf577113889b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:42:43.088407Z","signature_b64":"QDvs4c2wdCyBNUh6plvNQRg5yl+xJiO7oC/P+CC8CY7imbCJP9LWsm4RzmJD+CPzHPYl8lw91vMRNBaqylI2Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"954c36973d48de2b85b80a71f9c59827b6386b59b7564d307d951de52b28f11b","last_reissued_at":"2026-07-05T06:42:43.087916Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:42:43.087916Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Zero-Shot Composed Image Retrieval with Textual Inversion","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.IR"],"primary_cat":"cs.CV","authors_text":"Alberto Baldrati, Alberto Del Bimbo, Lorenzo Agnolucci, Marco Bertini","submitted_at":"2023-03-27T14:31:25Z","abstract_excerpt":"Composed Image Retrieval (CIR) aims to retrieve a target image based on a query composed of a reference image and a relative caption that describes the difference between the two images. The high effort and cost required for labeling datasets for CIR hamper the widespread usage of existing methods, as they rely on supervised learning. In this work, we propose a new task, Zero-Shot CIR (ZS-CIR), that aims to address CIR without requiring a labeled training dataset. Our approach, named zero-Shot composEd imAge Retrieval with textuaL invErsion (SEARLE), maps the visual features of the reference i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.15247","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/2303.15247/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":"2303.15247","created_at":"2026-07-05T06:42:43.087973+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.15247v2","created_at":"2026-07-05T06:42:43.087973+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.15247","created_at":"2026-07-05T06:42:43.087973+00:00"},{"alias_kind":"pith_short_12","alias_value":"SVGDNFZ5JDPC","created_at":"2026-07-05T06:42:43.087973+00:00"},{"alias_kind":"pith_short_16","alias_value":"SVGDNFZ5JDPCXBNY","created_at":"2026-07-05T06:42:43.087973+00:00"},{"alias_kind":"pith_short_8","alias_value":"SVGDNFZ5","created_at":"2026-07-05T06:42:43.087973+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07032","citing_title":"Never Seen Before: Benchmarking Genuine Zero-Shot Composed Image Retrieval with Consistent Video-Sourced Datasets","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6","json":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6.json","graph_json":"https://pith.science/api/pith-number/SVGDNFZ5JDPCXBNYBJY7TRMYE6/graph.json","events_json":"https://pith.science/api/pith-number/SVGDNFZ5JDPCXBNYBJY7TRMYE6/events.json","paper":"https://pith.science/paper/SVGDNFZ5"},"agent_actions":{"view_html":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6","download_json":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6.json","view_paper":"https://pith.science/paper/SVGDNFZ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.15247&json=true","fetch_graph":"https://pith.science/api/pith-number/SVGDNFZ5JDPCXBNYBJY7TRMYE6/graph.json","fetch_events":"https://pith.science/api/pith-number/SVGDNFZ5JDPCXBNYBJY7TRMYE6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6/action/storage_attestation","attest_author":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6/action/author_attestation","sign_citation":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6/action/citation_signature","submit_replication":"https://pith.science/pith/SVGDNFZ5JDPCXBNYBJY7TRMYE6/action/replication_record"}},"created_at":"2026-07-05T06:42:43.087973+00:00","updated_at":"2026-07-05T06:42:43.087973+00:00"}