{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XRXWECVFLGV4Q5LJFREF6PO7BI","short_pith_number":"pith:XRXWECVF","schema_version":"1.0","canonical_sha256":"bc6f620aa559abc875692c485f3ddf0a2166447be8bb2f2c235945a1190c996c","source":{"kind":"arxiv","id":"2503.17109","version":2},"attestation_state":"computed","paper":{"title":"Missing Target-Relevant Information Prediction with World Model for Accurate Zero-Shot Composed Image Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gang Xiong, Gaopeng Gou, Jiamin Zhuang, Jing Yu, Keke Gai, Qi Wu, Yuanmin Tang","submitted_at":"2025-03-21T12:49:50Z","abstract_excerpt":"Zero-Shot Composed Image Retrieval (ZS-CIR) involves diverse tasks with a broad range of visual content manipulation intent across domain, scene, object, and attribute. The key challenge for ZS-CIR tasks is to modify a reference image according to manipulation text to accurately retrieve a target image, especially when the reference image is missing essential target content. In this paper, we propose a novel prediction-based mapping network, named PrediCIR, to adaptively predict the missing target visual content in reference images in the latent space before mapping for accurate ZS-CIR. Specif"},"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":"2503.17109","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-03-21T12:49:50Z","cross_cats_sorted":[],"title_canon_sha256":"e8643ea19931565de244c00e81c3fc9e6e2405846b424461dc29029d00241c47","abstract_canon_sha256":"83c83056fcb79d7583e22ae8e9fd99e167fd5777466d6cd5650c06da25a73078"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:41:22.114904Z","signature_b64":"qJTfyE2fTHNFIyv0DSrFvEMulsFBUv2C6WkH0rgW33azIUX/bS8Zck21HJmOHAsrixmrAilkgh/7TknPEJ4iCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc6f620aa559abc875692c485f3ddf0a2166447be8bb2f2c235945a1190c996c","last_reissued_at":"2026-07-05T10:41:22.114398Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:41:22.114398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Missing Target-Relevant Information Prediction with World Model for Accurate Zero-Shot Composed Image Retrieval","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gang Xiong, Gaopeng Gou, Jiamin Zhuang, Jing Yu, Keke Gai, Qi Wu, Yuanmin Tang","submitted_at":"2025-03-21T12:49:50Z","abstract_excerpt":"Zero-Shot Composed Image Retrieval (ZS-CIR) involves diverse tasks with a broad range of visual content manipulation intent across domain, scene, object, and attribute. The key challenge for ZS-CIR tasks is to modify a reference image according to manipulation text to accurately retrieve a target image, especially when the reference image is missing essential target content. In this paper, we propose a novel prediction-based mapping network, named PrediCIR, to adaptively predict the missing target visual content in reference images in the latent space before mapping for accurate ZS-CIR. Specif"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.17109","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/2503.17109/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":"2503.17109","created_at":"2026-07-05T10:41:22.114460+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.17109v2","created_at":"2026-07-05T10:41:22.114460+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.17109","created_at":"2026-07-05T10:41:22.114460+00:00"},{"alias_kind":"pith_short_12","alias_value":"XRXWECVFLGV4","created_at":"2026-07-05T10:41:22.114460+00:00"},{"alias_kind":"pith_short_16","alias_value":"XRXWECVFLGV4Q5LJ","created_at":"2026-07-05T10:41:22.114460+00:00"},{"alias_kind":"pith_short_8","alias_value":"XRXWECVF","created_at":"2026-07-05T10:41:22.114460+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.14710","citing_title":"G-MIXER: Geodesic Mixup-based Implicit Semantic Expansion and Explicit Semantic Re-ranking for Zero-Shot Composed Image Retrieval","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI","json":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI.json","graph_json":"https://pith.science/api/pith-number/XRXWECVFLGV4Q5LJFREF6PO7BI/graph.json","events_json":"https://pith.science/api/pith-number/XRXWECVFLGV4Q5LJFREF6PO7BI/events.json","paper":"https://pith.science/paper/XRXWECVF"},"agent_actions":{"view_html":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI","download_json":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI.json","view_paper":"https://pith.science/paper/XRXWECVF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.17109&json=true","fetch_graph":"https://pith.science/api/pith-number/XRXWECVFLGV4Q5LJFREF6PO7BI/graph.json","fetch_events":"https://pith.science/api/pith-number/XRXWECVFLGV4Q5LJFREF6PO7BI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI/action/storage_attestation","attest_author":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI/action/author_attestation","sign_citation":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI/action/citation_signature","submit_replication":"https://pith.science/pith/XRXWECVFLGV4Q5LJFREF6PO7BI/action/replication_record"}},"created_at":"2026-07-05T10:41:22.114460+00:00","updated_at":"2026-07-05T10:41:22.114460+00:00"}