{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:JDTYGEJBIRDG6T5BRNMHZRTUYT","short_pith_number":"pith:JDTYGEJB","schema_version":"1.0","canonical_sha256":"48e783112144466f4fa18b587cc674c4def6fbeb0ad52c01a462d25e8e3ce254","source":{"kind":"arxiv","id":"2206.02574","version":3},"attestation_state":"computed","paper":{"title":"On the duality between contrastive and non-contrastive self-supervised learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Adrien Bardes (FAIR, CIMS), Laurent Najman (LIGM), LIGM), Quentin Garrido (FAIR, WILLOW), Yann Lecun (FAIR, Yubei Chen (FAIR)","submitted_at":"2022-06-03T08:04:12Z","abstract_excerpt":"Recent approaches in self-supervised learning of image representations can be categorized into different families of methods and, in particular, can be divided into contrastive and non-contrastive approaches. While differences between the two families have been thoroughly discussed to motivate new approaches, we focus more on the theoretical similarities between them. By designing contrastive and covariance based non-contrastive criteria that can be related algebraically and shown to be equivalent under limited assumptions, we show how close those families can be. We further study popular meth"},"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":"2206.02574","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-03T08:04:12Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"619f3d2a7b2fa437b2b845dc46949c953653ccd1f120a190f2a34893eacde870","abstract_canon_sha256":"e7f82b11f44dc4afb82d93de6a433ed65ea9f49e0e1f0a2a86389508c671f650"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:24:23.963857Z","signature_b64":"X/SVPvvyC9J8TvQB4b56VqS2KiqfkhZXwglki3uP8bm5Sib6PIitO9RfOQTYkXrAQQ4N/Ida0L7slp7PxuIwBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"48e783112144466f4fa18b587cc674c4def6fbeb0ad52c01a462d25e8e3ce254","last_reissued_at":"2026-07-05T06:24:23.963372Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:24:23.963372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the duality between contrastive and non-contrastive self-supervised learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Adrien Bardes (FAIR, CIMS), Laurent Najman (LIGM), LIGM), Quentin Garrido (FAIR, WILLOW), Yann Lecun (FAIR, Yubei Chen (FAIR)","submitted_at":"2022-06-03T08:04:12Z","abstract_excerpt":"Recent approaches in self-supervised learning of image representations can be categorized into different families of methods and, in particular, can be divided into contrastive and non-contrastive approaches. While differences between the two families have been thoroughly discussed to motivate new approaches, we focus more on the theoretical similarities between them. By designing contrastive and covariance based non-contrastive criteria that can be related algebraically and shown to be equivalent under limited assumptions, we show how close those families can be. We further study popular meth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.02574","kind":"arxiv","version":3},"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/2206.02574/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":"2206.02574","created_at":"2026-07-05T06:24:23.963431+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.02574v3","created_at":"2026-07-05T06:24:23.963431+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.02574","created_at":"2026-07-05T06:24:23.963431+00:00"},{"alias_kind":"pith_short_12","alias_value":"JDTYGEJBIRDG","created_at":"2026-07-05T06:24:23.963431+00:00"},{"alias_kind":"pith_short_16","alias_value":"JDTYGEJBIRDG6T5B","created_at":"2026-07-05T06:24:23.963431+00:00"},{"alias_kind":"pith_short_8","alias_value":"JDTYGEJB","created_at":"2026-07-05T06:24:23.963431+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":104,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":104,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18557","citing_title":"Self-supervised local learning rules learn the hidden hierarchical structure of high-dimensional data","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15394","citing_title":"Representation Without Reward: A JEPA Audit for LLM Fine-Tuning","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2510.26782","citing_title":"Cloning Deterministic Worlds: The Critical Role of Latent Geometry in Long-Horizon World Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2404.08471","citing_title":"Revisiting Feature Prediction for Learning Visual Representations from Video","ref_index":133,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT","json":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT.json","graph_json":"https://pith.science/api/pith-number/JDTYGEJBIRDG6T5BRNMHZRTUYT/graph.json","events_json":"https://pith.science/api/pith-number/JDTYGEJBIRDG6T5BRNMHZRTUYT/events.json","paper":"https://pith.science/paper/JDTYGEJB"},"agent_actions":{"view_html":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT","download_json":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT.json","view_paper":"https://pith.science/paper/JDTYGEJB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.02574&json=true","fetch_graph":"https://pith.science/api/pith-number/JDTYGEJBIRDG6T5BRNMHZRTUYT/graph.json","fetch_events":"https://pith.science/api/pith-number/JDTYGEJBIRDG6T5BRNMHZRTUYT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT/action/storage_attestation","attest_author":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT/action/author_attestation","sign_citation":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT/action/citation_signature","submit_replication":"https://pith.science/pith/JDTYGEJBIRDG6T5BRNMHZRTUYT/action/replication_record"}},"created_at":"2026-07-05T06:24:23.963431+00:00","updated_at":"2026-07-05T06:24:23.963431+00:00"}