{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:JUXDGH2G3MAHCZW3IHJYUL3LDS","short_pith_number":"pith:JUXDGH2G","schema_version":"1.0","canonical_sha256":"4d2e331f46db007166db41d38a2f6b1c89e01b9bd1c139c2a83a9cb687fc7cb8","source":{"kind":"arxiv","id":"2311.17049","version":2},"attestation_state":"computed","paper":{"title":"MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Fartash Faghri, Hadi Pouransari, Oncel Tuzel, Pavan Kumar Anasosalu Vasu, Raviteja Vemulapalli","submitted_at":"2023-11-28T18:55:42Z","abstract_excerpt":"Contrastive pretraining of image-text foundation models, such as CLIP, demonstrated excellent zero-shot performance and improved robustness on a wide range of downstream tasks. However, these models utilize large transformer-based encoders with significant memory and latency overhead which pose challenges for deployment on mobile devices. In this work, we introduce MobileCLIP -- a new family of efficient image-text models optimized for runtime performance along with a novel and efficient training approach, namely multi-modal reinforced training. The proposed training approach leverages knowled"},"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":"2311.17049","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-11-28T18:55:42Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"33c76664002c96b6c1072073e8d44ea907c645def518448c5635ed8a0d52832a","abstract_canon_sha256":"2624bbf32d8b8687d09f57e65c20d7bb53bc3d848205f8d3666423ec2687ad69"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:48.625548Z","signature_b64":"2HnpAbiTexYDvSLd4qjWa2qUl/98x9HhhZXXVJ2yhtgKnz1U+xi0wqbLIflabLUcVwQKxADXofX+KhpKWywdBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4d2e331f46db007166db41d38a2f6b1c89e01b9bd1c139c2a83a9cb687fc7cb8","last_reissued_at":"2026-07-05T08:02:48.624793Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:48.624793Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Fartash Faghri, Hadi Pouransari, Oncel Tuzel, Pavan Kumar Anasosalu Vasu, Raviteja Vemulapalli","submitted_at":"2023-11-28T18:55:42Z","abstract_excerpt":"Contrastive pretraining of image-text foundation models, such as CLIP, demonstrated excellent zero-shot performance and improved robustness on a wide range of downstream tasks. However, these models utilize large transformer-based encoders with significant memory and latency overhead which pose challenges for deployment on mobile devices. In this work, we introduce MobileCLIP -- a new family of efficient image-text models optimized for runtime performance along with a novel and efficient training approach, namely multi-modal reinforced training. The proposed training approach leverages knowled"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.17049","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/2311.17049/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":"2311.17049","created_at":"2026-07-05T08:02:48.624888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.17049v2","created_at":"2026-07-05T08:02:48.624888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.17049","created_at":"2026-07-05T08:02:48.624888+00:00"},{"alias_kind":"pith_short_12","alias_value":"JUXDGH2G3MAH","created_at":"2026-07-05T08:02:48.624888+00:00"},{"alias_kind":"pith_short_16","alias_value":"JUXDGH2G3MAHCZW3","created_at":"2026-07-05T08:02:48.624888+00:00"},{"alias_kind":"pith_short_8","alias_value":"JUXDGH2G","created_at":"2026-07-05T08:02:48.624888+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24253","citing_title":"TuringViT: Making SOTA Vision Transformers Accessible to All","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02371","citing_title":"VisionAId: An Offline-First Multimodal Android Assistant for People with Visual Impairment, Featuring Personalized Object Retrieval","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24253","citing_title":"TuringViT: Making SOTA Vision Transformers Accessible to All","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19054","citing_title":"Evaluation of Winning Solutions of 2025 Low Power Computer Vision Challenge","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS","json":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS.json","graph_json":"https://pith.science/api/pith-number/JUXDGH2G3MAHCZW3IHJYUL3LDS/graph.json","events_json":"https://pith.science/api/pith-number/JUXDGH2G3MAHCZW3IHJYUL3LDS/events.json","paper":"https://pith.science/paper/JUXDGH2G"},"agent_actions":{"view_html":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS","download_json":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS.json","view_paper":"https://pith.science/paper/JUXDGH2G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.17049&json=true","fetch_graph":"https://pith.science/api/pith-number/JUXDGH2G3MAHCZW3IHJYUL3LDS/graph.json","fetch_events":"https://pith.science/api/pith-number/JUXDGH2G3MAHCZW3IHJYUL3LDS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS/action/storage_attestation","attest_author":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS/action/author_attestation","sign_citation":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS/action/citation_signature","submit_replication":"https://pith.science/pith/JUXDGH2G3MAHCZW3IHJYUL3LDS/action/replication_record"}},"created_at":"2026-07-05T08:02:48.624888+00:00","updated_at":"2026-07-05T08:02:48.624888+00:00"}