{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XY2FZJZLAAPUQDTLSO7M55NOVN","short_pith_number":"pith:XY2FZJZL","schema_version":"1.0","canonical_sha256":"be345ca72b001f480e6b93becef5aeab4edcb61218d10404b37bfdf27297f2b5","source":{"kind":"arxiv","id":"2508.04424","version":3},"attestation_state":"computed","paper":{"title":"Composed Object Retrieval: Object-level Retrieval via Composed Expressions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fahad Shahbaz Khan, Guanyu Yang, Jinxing Zhou, Nian Liu, Salman Khan, Tong Wang, Zongyan Han","submitted_at":"2025-08-06T13:11:40Z","abstract_excerpt":"Retrieving fine-grained visual content based on user intent remains a challenge in multimodal systems. Although current Composed Image Retrieval (CIR) methods combine reference images with retrieval texts, they are constrained to image-level matching and cannot localize specific objects. To this end, we propose Composed Object Retrieval (COR), a new object-level retrieval task that retrieves target object(s) from candidate objects in a target image and grounds the retrieved result with pixel-level masks. Given a reference object, its mask, a target image, and a retrieval text describing the de"},"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":"2508.04424","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-06T13:11:40Z","cross_cats_sorted":[],"title_canon_sha256":"231cfc607e17f4f8761cb09e1c4265cb2902d9aa059ed5bd7fc019e06be03195","abstract_canon_sha256":"be427863419199cd386b25d423e89a95d56a59ca767fdeddd7f9dea672399a07"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:47.169716Z","signature_b64":"SM0hFWqy5B3KdGxDmH3dv7gTFvwcAO9D78s9sQ0EssFyotUSHCnM2ACT96NNjikaPF0nBsTdaaPdsHYp4eszAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be345ca72b001f480e6b93becef5aeab4edcb61218d10404b37bfdf27297f2b5","last_reissued_at":"2026-06-19T16:12:47.169232Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:47.169232Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Composed Object Retrieval: Object-level Retrieval via Composed Expressions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fahad Shahbaz Khan, Guanyu Yang, Jinxing Zhou, Nian Liu, Salman Khan, Tong Wang, Zongyan Han","submitted_at":"2025-08-06T13:11:40Z","abstract_excerpt":"Retrieving fine-grained visual content based on user intent remains a challenge in multimodal systems. Although current Composed Image Retrieval (CIR) methods combine reference images with retrieval texts, they are constrained to image-level matching and cannot localize specific objects. To this end, we propose Composed Object Retrieval (COR), a new object-level retrieval task that retrieves target object(s) from candidate objects in a target image and grounds the retrieved result with pixel-level masks. Given a reference object, its mask, a target image, and a retrieval text describing the de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.04424","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/2508.04424/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":"2508.04424","created_at":"2026-06-19T16:12:47.169293+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.04424v3","created_at":"2026-06-19T16:12:47.169293+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.04424","created_at":"2026-06-19T16:12:47.169293+00:00"},{"alias_kind":"pith_short_12","alias_value":"XY2FZJZLAAPU","created_at":"2026-06-19T16:12:47.169293+00:00"},{"alias_kind":"pith_short_16","alias_value":"XY2FZJZLAAPUQDTL","created_at":"2026-06-19T16:12:47.169293+00:00"},{"alias_kind":"pith_short_8","alias_value":"XY2FZJZL","created_at":"2026-06-19T16:12:47.169293+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN","json":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN.json","graph_json":"https://pith.science/api/pith-number/XY2FZJZLAAPUQDTLSO7M55NOVN/graph.json","events_json":"https://pith.science/api/pith-number/XY2FZJZLAAPUQDTLSO7M55NOVN/events.json","paper":"https://pith.science/paper/XY2FZJZL"},"agent_actions":{"view_html":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN","download_json":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN.json","view_paper":"https://pith.science/paper/XY2FZJZL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.04424&json=true","fetch_graph":"https://pith.science/api/pith-number/XY2FZJZLAAPUQDTLSO7M55NOVN/graph.json","fetch_events":"https://pith.science/api/pith-number/XY2FZJZLAAPUQDTLSO7M55NOVN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN/action/storage_attestation","attest_author":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN/action/author_attestation","sign_citation":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN/action/citation_signature","submit_replication":"https://pith.science/pith/XY2FZJZLAAPUQDTLSO7M55NOVN/action/replication_record"}},"created_at":"2026-06-19T16:12:47.169293+00:00","updated_at":"2026-06-19T16:12:47.169293+00:00"}