{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:X2HSXLQ36OP2P7NDLGORIIMWFW","short_pith_number":"pith:X2HSXLQ3","schema_version":"1.0","canonical_sha256":"be8f2bae1bf39fa7fda3599d1421962db9df0456c0b05648694b30b3174fffc1","source":{"kind":"arxiv","id":"2103.10957","version":2},"attestation_state":"computed","paper":{"title":"Efficient Visual Pretraining with Contrastive Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aaron van den Oord, Jean-Baptiste Alayrac, Jo\\~ao Carreira, Olivier J. H\\'enaff, Oriol Vinyals, Skanda Koppula","submitted_at":"2021-03-19T14:05:12Z","abstract_excerpt":"Self-supervised pretraining has been shown to yield powerful representations for transfer learning. These performance gains come at a large computational cost however, with state-of-the-art methods requiring an order of magnitude more computation than supervised pretraining. We tackle this computational bottleneck by introducing a new self-supervised objective, contrastive detection, which tasks representations with identifying object-level features across augmentations. This objective extracts a rich learning signal per image, leading to state-of-the-art transfer accuracy on a variety of down"},"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":"2103.10957","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-03-19T14:05:12Z","cross_cats_sorted":[],"title_canon_sha256":"3253129bdc70170876bebdd0680ed0e1f949d35bfb37a507a091a9e275de96f9","abstract_canon_sha256":"b899fa8282e78c7c26bad6fb5ea549bf73cae3a4b5a8c0fd68bf421fe15ac010"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:03:27.706605Z","signature_b64":"4y06pdwmDhbv/Oh9Y24o8bld+YFf2wk2mJXBskIu5aHlmrTYXSa6KYolCi9aa8pwLx677anMqM7abEgeVydMCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be8f2bae1bf39fa7fda3599d1421962db9df0456c0b05648694b30b3174fffc1","last_reissued_at":"2026-07-05T03:03:27.706065Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:03:27.706065Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Visual Pretraining with Contrastive Detection","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aaron van den Oord, Jean-Baptiste Alayrac, Jo\\~ao Carreira, Olivier J. H\\'enaff, Oriol Vinyals, Skanda Koppula","submitted_at":"2021-03-19T14:05:12Z","abstract_excerpt":"Self-supervised pretraining has been shown to yield powerful representations for transfer learning. These performance gains come at a large computational cost however, with state-of-the-art methods requiring an order of magnitude more computation than supervised pretraining. We tackle this computational bottleneck by introducing a new self-supervised objective, contrastive detection, which tasks representations with identifying object-level features across augmentations. This objective extracts a rich learning signal per image, leading to state-of-the-art transfer accuracy on a variety of down"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.10957","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/2103.10957/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":"2103.10957","created_at":"2026-07-05T03:03:27.706125+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.10957v2","created_at":"2026-07-05T03:03:27.706125+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.10957","created_at":"2026-07-05T03:03:27.706125+00:00"},{"alias_kind":"pith_short_12","alias_value":"X2HSXLQ36OP2","created_at":"2026-07-05T03:03:27.706125+00:00"},{"alias_kind":"pith_short_16","alias_value":"X2HSXLQ36OP2P7ND","created_at":"2026-07-05T03:03:27.706125+00:00"},{"alias_kind":"pith_short_8","alias_value":"X2HSXLQ3","created_at":"2026-07-05T03:03:27.706125+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2309.16588","citing_title":"Vision Transformers Need Registers","ref_index":87,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW","json":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW.json","graph_json":"https://pith.science/api/pith-number/X2HSXLQ36OP2P7NDLGORIIMWFW/graph.json","events_json":"https://pith.science/api/pith-number/X2HSXLQ36OP2P7NDLGORIIMWFW/events.json","paper":"https://pith.science/paper/X2HSXLQ3"},"agent_actions":{"view_html":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW","download_json":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW.json","view_paper":"https://pith.science/paper/X2HSXLQ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.10957&json=true","fetch_graph":"https://pith.science/api/pith-number/X2HSXLQ36OP2P7NDLGORIIMWFW/graph.json","fetch_events":"https://pith.science/api/pith-number/X2HSXLQ36OP2P7NDLGORIIMWFW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW/action/storage_attestation","attest_author":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW/action/author_attestation","sign_citation":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW/action/citation_signature","submit_replication":"https://pith.science/pith/X2HSXLQ36OP2P7NDLGORIIMWFW/action/replication_record"}},"created_at":"2026-07-05T03:03:27.706125+00:00","updated_at":"2026-07-05T03:03:27.706125+00:00"}