{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XHUYMOM4OMSPKFYNXJ45ZCMHKF","short_pith_number":"pith:XHUYMOM4","schema_version":"1.0","canonical_sha256":"b9e986399c7324f5170dba79dc89875156976343cc54c238070ee7e77fd5c564","source":{"kind":"arxiv","id":"2502.20826","version":1},"attestation_state":"computed","paper":{"title":"CoTMR: Chain-of-Thought Multi-Scale Reasoning for Training-Free Zero-Shot Composed Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CV","authors_text":"Dong Jing, Zelong Sun, Zhiwu Lu","submitted_at":"2025-02-28T08:12:23Z","abstract_excerpt":"Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images by integrating information from a composed query (reference image and modification text) without training samples. Existing methods primarily combine caption models and large language models (LLMs) to generate target captions based on composed queries but face various issues such as incompatibility, visual information loss, and insufficient reasoning. In this work, we propose CoTMR, a training-free framework crafted for ZS-CIR with novel Chain-of-thought (CoT) and Multi-scale Reasoning. Instead of relying on caption mod"},"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":"2502.20826","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-28T08:12:23Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"46e4b2a869e28bb3d486965089576111d2bbf33bdc9147493d9b8b440eaea9b1","abstract_canon_sha256":"60adf3ac1f2c62855fc1606b1948ca36ca80509ee8998c26b23197c311aa1278"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:40.469780Z","signature_b64":"vwS5/GgGZDPLifKSY/hftV/9v+H5qtwKI0E9VOlcU6Cy2oj52zL7AoSBO3viVH0CnZZ6nZ56j07KZoRKVoe6DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b9e986399c7324f5170dba79dc89875156976343cc54c238070ee7e77fd5c564","last_reissued_at":"2026-07-05T10:21:40.469277Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:40.469277Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoTMR: Chain-of-Thought Multi-Scale Reasoning for Training-Free Zero-Shot Composed Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CV","authors_text":"Dong Jing, Zelong Sun, Zhiwu Lu","submitted_at":"2025-02-28T08:12:23Z","abstract_excerpt":"Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images by integrating information from a composed query (reference image and modification text) without training samples. Existing methods primarily combine caption models and large language models (LLMs) to generate target captions based on composed queries but face various issues such as incompatibility, visual information loss, and insufficient reasoning. In this work, we propose CoTMR, a training-free framework crafted for ZS-CIR with novel Chain-of-thought (CoT) and Multi-scale Reasoning. Instead of relying on caption mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20826","kind":"arxiv","version":1},"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/2502.20826/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":"2502.20826","created_at":"2026-07-05T10:21:40.469340+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20826v1","created_at":"2026-07-05T10:21:40.469340+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20826","created_at":"2026-07-05T10:21:40.469340+00:00"},{"alias_kind":"pith_short_12","alias_value":"XHUYMOM4OMSP","created_at":"2026-07-05T10:21:40.469340+00:00"},{"alias_kind":"pith_short_16","alias_value":"XHUYMOM4OMSPKFYN","created_at":"2026-07-05T10:21:40.469340+00:00"},{"alias_kind":"pith_short_8","alias_value":"XHUYMOM4","created_at":"2026-07-05T10:21:40.469340+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00910","citing_title":"Reason, Retrieve, Re-rank: A Zero-Shot Reasoning-Aware Framework for Composed Video Retrieval","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF","json":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF.json","graph_json":"https://pith.science/api/pith-number/XHUYMOM4OMSPKFYNXJ45ZCMHKF/graph.json","events_json":"https://pith.science/api/pith-number/XHUYMOM4OMSPKFYNXJ45ZCMHKF/events.json","paper":"https://pith.science/paper/XHUYMOM4"},"agent_actions":{"view_html":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF","download_json":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF.json","view_paper":"https://pith.science/paper/XHUYMOM4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20826&json=true","fetch_graph":"https://pith.science/api/pith-number/XHUYMOM4OMSPKFYNXJ45ZCMHKF/graph.json","fetch_events":"https://pith.science/api/pith-number/XHUYMOM4OMSPKFYNXJ45ZCMHKF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF/action/storage_attestation","attest_author":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF/action/author_attestation","sign_citation":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF/action/citation_signature","submit_replication":"https://pith.science/pith/XHUYMOM4OMSPKFYNXJ45ZCMHKF/action/replication_record"}},"created_at":"2026-07-05T10:21:40.469340+00:00","updated_at":"2026-07-05T10:21:40.469340+00:00"}