{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HJVE2A7IFTGQ3AI5XCKNGJE2V2","short_pith_number":"pith:HJVE2A7I","schema_version":"1.0","canonical_sha256":"3a6a4d03e82ccd0d811db894d3249aae9914f4e9fb0053c6073f9e40bba9d1b6","source":{"kind":"arxiv","id":"2110.05208","version":2},"attestation_state":"computed","paper":{"title":"Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Liang, Fengwei Yu, Jing Shao, Junjie Yan, Lichen Zhao, Wanli Ouyang, Yangguang Li, Yufeng Cui","submitted_at":"2021-10-11T12:17:32Z","abstract_excerpt":"Recently, large-scale Contrastive Language-Image Pre-training (CLIP) has attracted unprecedented attention for its impressive zero-shot recognition ability and excellent transferability to downstream tasks. However, CLIP is quite data-hungry and requires 400M image-text pairs for pre-training, thereby restricting its adoption. This work proposes a novel training paradigm, Data efficient CLIP (DeCLIP), to alleviate this limitation. We demonstrate that by carefully utilizing the widespread supervision among the image-text pairs, our De-CLIP can learn generic visual features more efficiently. Ins"},"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":"2110.05208","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-10-11T12:17:32Z","cross_cats_sorted":[],"title_canon_sha256":"9b822f38128a3688bdacc6c46316b7fc00686c4273cc2dd626ca640a8c652007","abstract_canon_sha256":"5286545f06eda2c0884398ca88e0ebf98b8104074f9a45c05011167ee58aaa1e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:04:42.238660Z","signature_b64":"5WSm7oe1qaToRXIRsJrzVhAEfo3k+NuprgqEWU816vWxQbvE1LqSq807/AzN8IXHDaMqj6lLnyP5AikfjBWpDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a6a4d03e82ccd0d811db894d3249aae9914f4e9fb0053c6073f9e40bba9d1b6","last_reissued_at":"2026-07-05T04:04:42.238216Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:04:42.238216Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Supervision Exists Everywhere: A Data Efficient Contrastive Language-Image Pre-training Paradigm","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Liang, Fengwei Yu, Jing Shao, Junjie Yan, Lichen Zhao, Wanli Ouyang, Yangguang Li, Yufeng Cui","submitted_at":"2021-10-11T12:17:32Z","abstract_excerpt":"Recently, large-scale Contrastive Language-Image Pre-training (CLIP) has attracted unprecedented attention for its impressive zero-shot recognition ability and excellent transferability to downstream tasks. However, CLIP is quite data-hungry and requires 400M image-text pairs for pre-training, thereby restricting its adoption. This work proposes a novel training paradigm, Data efficient CLIP (DeCLIP), to alleviate this limitation. We demonstrate that by carefully utilizing the widespread supervision among the image-text pairs, our De-CLIP can learn generic visual features more efficiently. Ins"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.05208","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/2110.05208/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":"2110.05208","created_at":"2026-07-05T04:04:42.238271+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.05208v2","created_at":"2026-07-05T04:04:42.238271+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.05208","created_at":"2026-07-05T04:04:42.238271+00:00"},{"alias_kind":"pith_short_12","alias_value":"HJVE2A7IFTGQ","created_at":"2026-07-05T04:04:42.238271+00:00"},{"alias_kind":"pith_short_16","alias_value":"HJVE2A7IFTGQ3AI5","created_at":"2026-07-05T04:04:42.238271+00:00"},{"alias_kind":"pith_short_8","alias_value":"HJVE2A7I","created_at":"2026-07-05T04:04:42.238271+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07907","citing_title":"Multimodal Unlearning Across Vision, Language, Video, and Audio: Survey of Methods, Datasets, and Benchmarks","ref_index":43,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26794","citing_title":"ReasonCLIP-58M: Visually Grounded Commonsense Reasoning Supervision for CLIP","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2208.14649","citing_title":"DetailCLIP: Injecting Image Details into CLIP's Feature Space","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15615","citing_title":"Neutral-Reference Prompting for Vision-Language Models","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16671","citing_title":"Demystifying CLIP Data","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2403.14608","citing_title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","ref_index":201,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13970","citing_title":"MApLe: Multi-instance Alignment of Diagnostic Reports and Large Medical Images","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2","json":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2.json","graph_json":"https://pith.science/api/pith-number/HJVE2A7IFTGQ3AI5XCKNGJE2V2/graph.json","events_json":"https://pith.science/api/pith-number/HJVE2A7IFTGQ3AI5XCKNGJE2V2/events.json","paper":"https://pith.science/paper/HJVE2A7I"},"agent_actions":{"view_html":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2","download_json":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2.json","view_paper":"https://pith.science/paper/HJVE2A7I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.05208&json=true","fetch_graph":"https://pith.science/api/pith-number/HJVE2A7IFTGQ3AI5XCKNGJE2V2/graph.json","fetch_events":"https://pith.science/api/pith-number/HJVE2A7IFTGQ3AI5XCKNGJE2V2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2/action/storage_attestation","attest_author":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2/action/author_attestation","sign_citation":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2/action/citation_signature","submit_replication":"https://pith.science/pith/HJVE2A7IFTGQ3AI5XCKNGJE2V2/action/replication_record"}},"created_at":"2026-07-05T04:04:42.238271+00:00","updated_at":"2026-07-05T04:04:42.238271+00:00"}