{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:5UTGYOLASW6UBMTGHMJPTCRXMO","short_pith_number":"pith:5UTGYOLA","schema_version":"1.0","canonical_sha256":"ed266c396095bd40b2663b12f98a37638db94135e1e7b65ff328e9eebff026f7","source":{"kind":"arxiv","id":"2305.16304","version":3},"attestation_state":"computed","paper":{"title":"Candidate Set Re-ranking for Composed Image Retrieval with Dual Multi-modal Encoder","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.IR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Damien Teney, Stephen Gould, Weixuan Sun, Zheyuan Liu","submitted_at":"2023-05-25T17:56:24Z","abstract_excerpt":"Composed image retrieval aims to find an image that best matches a given multi-modal user query consisting of a reference image and text pair. Existing methods commonly pre-compute image embeddings over the entire corpus and compare these to a reference image embedding modified by the query text at test time. Such a pipeline is very efficient at test time since fast vector distances can be used to evaluate candidates, but modifying the reference image embedding guided only by a short textual description can be difficult, especially independent of potential candidates. An alternative approach 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":"2305.16304","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-25T17:56:24Z","cross_cats_sorted":["cs.IR","cs.LG"],"title_canon_sha256":"e02b00f3b33d4f43222f8a1b97c11785ad8494bf46538b795d4a937047ae9c08","abstract_canon_sha256":"f344479eb237b4ec46cb389a43e5241e8c2b39620faaecc0d91ae804e15fbe69"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:38:28.068549Z","signature_b64":"J/rkNvYPzMPdLL79OnRh18d6GU5+4KwVY988hc8QoFWsMUsJgReyR+Pf8UGaVpM1dWb0JsujebGxL7OrOWErAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed266c396095bd40b2663b12f98a37638db94135e1e7b65ff328e9eebff026f7","last_reissued_at":"2026-07-05T07:38:28.068071Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:38:28.068071Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Candidate Set Re-ranking for Composed Image Retrieval with Dual Multi-modal Encoder","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.IR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Damien Teney, Stephen Gould, Weixuan Sun, Zheyuan Liu","submitted_at":"2023-05-25T17:56:24Z","abstract_excerpt":"Composed image retrieval aims to find an image that best matches a given multi-modal user query consisting of a reference image and text pair. Existing methods commonly pre-compute image embeddings over the entire corpus and compare these to a reference image embedding modified by the query text at test time. Such a pipeline is very efficient at test time since fast vector distances can be used to evaluate candidates, but modifying the reference image embedding guided only by a short textual description can be difficult, especially independent of potential candidates. An alternative approach i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.16304","kind":"arxiv","version":3},"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/2305.16304/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":"2305.16304","created_at":"2026-07-05T07:38:28.068128+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.16304v3","created_at":"2026-07-05T07:38:28.068128+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.16304","created_at":"2026-07-05T07:38:28.068128+00:00"},{"alias_kind":"pith_short_12","alias_value":"5UTGYOLASW6U","created_at":"2026-07-05T07:38:28.068128+00:00"},{"alias_kind":"pith_short_16","alias_value":"5UTGYOLASW6UBMTG","created_at":"2026-07-05T07:38:28.068128+00:00"},{"alias_kind":"pith_short_8","alias_value":"5UTGYOLA","created_at":"2026-07-05T07:38:28.068128+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.16487","citing_title":"Geometry-Aware CLIP Retrieval via Local Cross-Modal Alignment and Steering","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05583","citing_title":"WRF4CIR: Weight-Regularized Fine-Tuning Network for Composed Image Retrieval","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO","json":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO.json","graph_json":"https://pith.science/api/pith-number/5UTGYOLASW6UBMTGHMJPTCRXMO/graph.json","events_json":"https://pith.science/api/pith-number/5UTGYOLASW6UBMTGHMJPTCRXMO/events.json","paper":"https://pith.science/paper/5UTGYOLA"},"agent_actions":{"view_html":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO","download_json":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO.json","view_paper":"https://pith.science/paper/5UTGYOLA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.16304&json=true","fetch_graph":"https://pith.science/api/pith-number/5UTGYOLASW6UBMTGHMJPTCRXMO/graph.json","fetch_events":"https://pith.science/api/pith-number/5UTGYOLASW6UBMTGHMJPTCRXMO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO/action/storage_attestation","attest_author":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO/action/author_attestation","sign_citation":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO/action/citation_signature","submit_replication":"https://pith.science/pith/5UTGYOLASW6UBMTGHMJPTCRXMO/action/replication_record"}},"created_at":"2026-07-05T07:38:28.068128+00:00","updated_at":"2026-07-05T07:38:28.068128+00:00"}