{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:JJJMBZNLRGTEXCBLE3LISUAXIE","short_pith_number":"pith:JJJMBZNL","schema_version":"1.0","canonical_sha256":"4a52c0e5ab89a64b882b26d689501741367b4e187cf66474187836ac51e10bcd","source":{"kind":"arxiv","id":"2403.19651","version":2},"attestation_state":"computed","paper":{"title":"MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR","cs.MM"],"primary_cat":"cs.CV","authors_text":"Hexiang Hu, Kai Zhang, Kenton Lee, Ming-Wei Chang, Siyuan Qiao, Wenhu Chen, Yi Luan, Yu Su","submitted_at":"2024-03-28T17:59:20Z","abstract_excerpt":"Image retrieval, i.e., finding desired images given a reference image, inherently encompasses rich, multi-faceted search intents that are difficult to capture solely using image-based measures. Recent works leverage text instructions to allow users to more freely express their search intents. However, they primarily focus on image pairs that are visually similar and/or can be characterized by a small set of pre-defined relations. The core thesis of this paper is that text instructions can enable retrieving images with richer relations beyond visual similarity. To show this, we introduce MagicL"},"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":"2403.19651","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-28T17:59:20Z","cross_cats_sorted":["cs.AI","cs.CL","cs.IR","cs.MM"],"title_canon_sha256":"721c344b695cfda73755df30c2df0184df2211da9250f989e76ce724882ec188","abstract_canon_sha256":"faf546a712bf18ff984da6cf312484e54d300a1fe597c211c4ef404f6c32d015"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:24.628752Z","signature_b64":"Unqds5/xvbKqQSFEARJ8eXDB4II+gEQgT79Xn+cnKrP1ULImcir64/PNDC8q4jKfWDHqPx4nExQEAPgUpDdzAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a52c0e5ab89a64b882b26d689501741367b4e187cf66474187836ac51e10bcd","last_reissued_at":"2026-07-05T08:36:24.628278Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:24.628278Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.IR","cs.MM"],"primary_cat":"cs.CV","authors_text":"Hexiang Hu, Kai Zhang, Kenton Lee, Ming-Wei Chang, Siyuan Qiao, Wenhu Chen, Yi Luan, Yu Su","submitted_at":"2024-03-28T17:59:20Z","abstract_excerpt":"Image retrieval, i.e., finding desired images given a reference image, inherently encompasses rich, multi-faceted search intents that are difficult to capture solely using image-based measures. Recent works leverage text instructions to allow users to more freely express their search intents. However, they primarily focus on image pairs that are visually similar and/or can be characterized by a small set of pre-defined relations. The core thesis of this paper is that text instructions can enable retrieving images with richer relations beyond visual similarity. To show this, we introduce MagicL"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.19651","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/2403.19651/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":"2403.19651","created_at":"2026-07-05T08:36:24.628354+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.19651v2","created_at":"2026-07-05T08:36:24.628354+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.19651","created_at":"2026-07-05T08:36:24.628354+00:00"},{"alias_kind":"pith_short_12","alias_value":"JJJMBZNLRGTE","created_at":"2026-07-05T08:36:24.628354+00:00"},{"alias_kind":"pith_short_16","alias_value":"JJJMBZNLRGTEXCBL","created_at":"2026-07-05T08:36:24.628354+00:00"},{"alias_kind":"pith_short_8","alias_value":"JJJMBZNL","created_at":"2026-07-05T08:36:24.628354+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19584","citing_title":"Language-Instructed Vision Embeddings for Controllable and Generalizable Perception","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21261","citing_title":"STiTch: Semantic Transition and Transportation in Collaboration for Training-Free Zero-Shot Composed Image Retrieval","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2410.05160","citing_title":"VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02073","citing_title":"PLUME: Latent Reasoning Based Universal Multimodal Embedding","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13710","citing_title":"SLQ: Bridging Modalities via Shared Latent Queries for Retrieval with Frozen MLLMs","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25273","citing_title":"Combating Visual Neglect and Semantic Drift in Large Multimodal Models for Enhanced Cross-Modal Retrieval","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13710","citing_title":"SLQ: Bridging Modalities via Shared Latent Queries for Retrieval with Frozen MLLMs","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE","json":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE.json","graph_json":"https://pith.science/api/pith-number/JJJMBZNLRGTEXCBLE3LISUAXIE/graph.json","events_json":"https://pith.science/api/pith-number/JJJMBZNLRGTEXCBLE3LISUAXIE/events.json","paper":"https://pith.science/paper/JJJMBZNL"},"agent_actions":{"view_html":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE","download_json":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE.json","view_paper":"https://pith.science/paper/JJJMBZNL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.19651&json=true","fetch_graph":"https://pith.science/api/pith-number/JJJMBZNLRGTEXCBLE3LISUAXIE/graph.json","fetch_events":"https://pith.science/api/pith-number/JJJMBZNLRGTEXCBLE3LISUAXIE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE/action/storage_attestation","attest_author":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE/action/author_attestation","sign_citation":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE/action/citation_signature","submit_replication":"https://pith.science/pith/JJJMBZNLRGTEXCBLE3LISUAXIE/action/replication_record"}},"created_at":"2026-07-05T08:36:24.628354+00:00","updated_at":"2026-07-05T08:36:24.628354+00:00"}