{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:XH3DQAMAD5R3YWDEDE2Q3ZBUUL","short_pith_number":"pith:XH3DQAMA","schema_version":"1.0","canonical_sha256":"b9f63801801f63bc586419350de434a2f846d2a52a46ea643604eac63811e7e8","source":{"kind":"arxiv","id":"2205.11508","version":3},"attestation_state":"computed","paper":{"title":"Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","math.SP","stat.ML"],"primary_cat":"cs.LG","authors_text":"Randall Balestriero, Yann LeCun","submitted_at":"2022-05-23T17:59:32Z","abstract_excerpt":"Self-Supervised Learning (SSL) surmises that inputs and pairwise positive relationships are enough to learn meaningful representations. Although SSL has recently reached a milestone: outperforming supervised methods in many modalities\\dots the theoretical foundations are limited, method-specific, and fail to provide principled design guidelines to practitioners. In this paper, we propose a unifying framework under the helm of spectral manifold learning to address those limitations. Through the course of this study, we will rigorously demonstrate that VICReg, SimCLR, BarlowTwins et al. correspo"},"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":"2205.11508","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-05-23T17:59:32Z","cross_cats_sorted":["cs.AI","cs.CV","math.SP","stat.ML"],"title_canon_sha256":"f42b77eb430c57a6a07000087ee46dfb28aa9c74abf6df35e7dbf235e1a98aa7","abstract_canon_sha256":"d473aeec816faa31f6832c450825ac013107efbf8eac84adc38fa4890abd3eb8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:30:42.195750Z","signature_b64":"FEWmk6nPjBRisDS1k4krPpOI3X1WNKzqZEFELHP5RDrvD52jLEnTLKGAZGx6yt/VpDZSWjuDbFEikMmFBNUQCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b9f63801801f63bc586419350de434a2f846d2a52a46ea643604eac63811e7e8","last_reissued_at":"2026-07-05T04:30:42.195305Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:30:42.195305Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrastive and Non-Contrastive Self-Supervised Learning Recover Global and Local Spectral Embedding Methods","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV","math.SP","stat.ML"],"primary_cat":"cs.LG","authors_text":"Randall Balestriero, Yann LeCun","submitted_at":"2022-05-23T17:59:32Z","abstract_excerpt":"Self-Supervised Learning (SSL) surmises that inputs and pairwise positive relationships are enough to learn meaningful representations. Although SSL has recently reached a milestone: outperforming supervised methods in many modalities\\dots the theoretical foundations are limited, method-specific, and fail to provide principled design guidelines to practitioners. In this paper, we propose a unifying framework under the helm of spectral manifold learning to address those limitations. Through the course of this study, we will rigorously demonstrate that VICReg, SimCLR, BarlowTwins et al. correspo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.11508","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/2205.11508/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":"2205.11508","created_at":"2026-07-05T04:30:42.195377+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.11508v3","created_at":"2026-07-05T04:30:42.195377+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.11508","created_at":"2026-07-05T04:30:42.195377+00:00"},{"alias_kind":"pith_short_12","alias_value":"XH3DQAMAD5R3","created_at":"2026-07-05T04:30:42.195377+00:00"},{"alias_kind":"pith_short_16","alias_value":"XH3DQAMAD5R3YWDE","created_at":"2026-07-05T04:30:42.195377+00:00"},{"alias_kind":"pith_short_8","alias_value":"XH3DQAMA","created_at":"2026-07-05T04:30:42.195377+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":106,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22873","citing_title":"SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning","ref_index":106,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26379","citing_title":"When Does LeJEPA Learn a World Model?","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2404.08471","citing_title":"Revisiting Feature Prediction for Learning Visual Representations from Video","ref_index":130,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL","json":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL.json","graph_json":"https://pith.science/api/pith-number/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/graph.json","events_json":"https://pith.science/api/pith-number/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/events.json","paper":"https://pith.science/paper/XH3DQAMA"},"agent_actions":{"view_html":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL","download_json":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL.json","view_paper":"https://pith.science/paper/XH3DQAMA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.11508&json=true","fetch_graph":"https://pith.science/api/pith-number/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/graph.json","fetch_events":"https://pith.science/api/pith-number/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/action/storage_attestation","attest_author":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/action/author_attestation","sign_citation":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/action/citation_signature","submit_replication":"https://pith.science/pith/XH3DQAMAD5R3YWDEDE2Q3ZBUUL/action/replication_record"}},"created_at":"2026-07-05T04:30:42.195377+00:00","updated_at":"2026-07-05T04:30:42.195377+00:00"}