{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:FV4MJEX24NRBIGCJJROXTZUHCG","short_pith_number":"pith:FV4MJEX2","schema_version":"1.0","canonical_sha256":"2d78c492fae3621418494c5d79e68711b818c2162843598acaacb5a3abd179a0","source":{"kind":"arxiv","id":"2306.07272","version":2},"attestation_state":"computed","paper":{"title":"Zero-shot Composed Text-Image Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiangchao Yao, Weidi Xie, Yanfeng Wang, Ya Zhang, Yikun Liu","submitted_at":"2023-06-12T17:56:01Z","abstract_excerpt":"In this paper, we consider the problem of composed image retrieval (CIR), it aims to train a model that can fuse multi-modal information, e.g., text and images, to accurately retrieve images that match the query, extending the user's expression ability. We make the following contributions: (i) we initiate a scalable pipeline to automatically construct datasets for training CIR model, by simply exploiting a large-scale dataset of image-text pairs, e.g., a subset of LAION-5B; (ii) we introduce a transformer-based adaptive aggregation model, TransAgg, which employs a simple yet efficient fusion m"},"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":"2306.07272","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-06-12T17:56:01Z","cross_cats_sorted":[],"title_canon_sha256":"83fdc52a574fb3fbc79accba51880af7f0da8465453946a3fef39210766a6462","abstract_canon_sha256":"e5d3ce53f4ea63afcbbe0b5574feba38710c609692978e4d3383042fe64406d1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:52:40.669408Z","signature_b64":"pqtRFpwPFK6K7BhxKjp9tRy3jkXAlspzhh8neNUNovv8WEutEZAHBnsUb1G1REDatRb9EFO+VKdezw58NxjfBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d78c492fae3621418494c5d79e68711b818c2162843598acaacb5a3abd179a0","last_reissued_at":"2026-07-05T07:52:40.668891Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:52:40.668891Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Zero-shot Composed Text-Image Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiangchao Yao, Weidi Xie, Yanfeng Wang, Ya Zhang, Yikun Liu","submitted_at":"2023-06-12T17:56:01Z","abstract_excerpt":"In this paper, we consider the problem of composed image retrieval (CIR), it aims to train a model that can fuse multi-modal information, e.g., text and images, to accurately retrieve images that match the query, extending the user's expression ability. We make the following contributions: (i) we initiate a scalable pipeline to automatically construct datasets for training CIR model, by simply exploiting a large-scale dataset of image-text pairs, e.g., a subset of LAION-5B; (ii) we introduce a transformer-based adaptive aggregation model, TransAgg, which employs a simple yet efficient fusion m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.07272","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/2306.07272/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":"2306.07272","created_at":"2026-07-05T07:52:40.668951+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.07272v2","created_at":"2026-07-05T07:52:40.668951+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.07272","created_at":"2026-07-05T07:52:40.668951+00:00"},{"alias_kind":"pith_short_12","alias_value":"FV4MJEX24NRB","created_at":"2026-07-05T07:52:40.668951+00:00"},{"alias_kind":"pith_short_16","alias_value":"FV4MJEX24NRBIGCJ","created_at":"2026-07-05T07:52:40.668951+00:00"},{"alias_kind":"pith_short_8","alias_value":"FV4MJEX2","created_at":"2026-07-05T07:52:40.668951+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"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":44,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03470","citing_title":"Mixed-Modality Dual Face-Hair Retrieval","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG","json":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG.json","graph_json":"https://pith.science/api/pith-number/FV4MJEX24NRBIGCJJROXTZUHCG/graph.json","events_json":"https://pith.science/api/pith-number/FV4MJEX24NRBIGCJJROXTZUHCG/events.json","paper":"https://pith.science/paper/FV4MJEX2"},"agent_actions":{"view_html":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG","download_json":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG.json","view_paper":"https://pith.science/paper/FV4MJEX2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.07272&json=true","fetch_graph":"https://pith.science/api/pith-number/FV4MJEX24NRBIGCJJROXTZUHCG/graph.json","fetch_events":"https://pith.science/api/pith-number/FV4MJEX24NRBIGCJJROXTZUHCG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG/action/storage_attestation","attest_author":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG/action/author_attestation","sign_citation":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG/action/citation_signature","submit_replication":"https://pith.science/pith/FV4MJEX24NRBIGCJJROXTZUHCG/action/replication_record"}},"created_at":"2026-07-05T07:52:40.668951+00:00","updated_at":"2026-07-05T07:52:40.668951+00:00"}