{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:XL42NLYEUVGILAWPYCDW3VTGTU","short_pith_number":"pith:XL42NLYE","schema_version":"1.0","canonical_sha256":"baf9a6af04a54c8582cfc0876dd6669d32cb85ce5cae4761f744171918b0d868","source":{"kind":"arxiv","id":"2210.07277","version":1},"attestation_state":"computed","paper":{"title":"The Hidden Uniform Cluster Prior in Self-Supervised Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Florian Bordes, Ishan Misra, Mahmoud Assran, Michael Rabbat, Nicolas Ballas, Pascal Vincent, Piotr Bojanowski, Quentin Duval, Randall Balestriero","submitted_at":"2022-10-13T18:10:01Z","abstract_excerpt":"A successful paradigm in representation learning is to perform self-supervised pretraining using tasks based on mini-batch statistics (e.g., SimCLR, VICReg, SwAV, MSN). We show that in the formulation of all these methods is an overlooked prior to learn features that enable uniform clustering of the data. While this prior has led to remarkably semantic representations when pretraining on class-balanced data, such as ImageNet, we demonstrate that it can hamper performance when pretraining on class-imbalanced data. By moving away from conventional uniformity priors and instead preferring power-l"},"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":"2210.07277","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-10-13T18:10:01Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"c11dcd15e04a6cf0372947d0ce86bb0389efbe6a9cf2a098862c1f2407856ef0","abstract_canon_sha256":"a41eaf2673b51399576ef00cf2f6d3634001975dfb434c15094a21fe30921e43"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:06:35.501938Z","signature_b64":"LiF4GIuVKKbT7M2BWXfhtAOTaC2imbiOkmryK7uKwwE2qx4HCUtmt/lRevi6MuWktDsjxApXTmHInoXi32BiAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"baf9a6af04a54c8582cfc0876dd6669d32cb85ce5cae4761f744171918b0d868","last_reissued_at":"2026-07-05T05:06:35.501592Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:06:35.501592Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Hidden Uniform Cluster Prior in Self-Supervised Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Florian Bordes, Ishan Misra, Mahmoud Assran, Michael Rabbat, Nicolas Ballas, Pascal Vincent, Piotr Bojanowski, Quentin Duval, Randall Balestriero","submitted_at":"2022-10-13T18:10:01Z","abstract_excerpt":"A successful paradigm in representation learning is to perform self-supervised pretraining using tasks based on mini-batch statistics (e.g., SimCLR, VICReg, SwAV, MSN). We show that in the formulation of all these methods is an overlooked prior to learn features that enable uniform clustering of the data. While this prior has led to remarkably semantic representations when pretraining on class-balanced data, such as ImageNet, we demonstrate that it can hamper performance when pretraining on class-imbalanced data. By moving away from conventional uniformity priors and instead preferring power-l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.07277","kind":"arxiv","version":1},"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/2210.07277/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":"2210.07277","created_at":"2026-07-05T05:06:35.501650+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.07277v1","created_at":"2026-07-05T05:06:35.501650+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.07277","created_at":"2026-07-05T05:06:35.501650+00:00"},{"alias_kind":"pith_short_12","alias_value":"XL42NLYEUVGI","created_at":"2026-07-05T05:06:35.501650+00:00"},{"alias_kind":"pith_short_16","alias_value":"XL42NLYEUVGILAWP","created_at":"2026-07-05T05:06:35.501650+00:00"},{"alias_kind":"pith_short_8","alias_value":"XL42NLYE","created_at":"2026-07-05T05:06:35.501650+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.08544","citing_title":"LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics","ref_index":92,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11870","citing_title":"Information theoretic underpinning of self-supervised learning by clustering","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2404.08471","citing_title":"Revisiting Feature Prediction for Learning Visual Representations from Video","ref_index":150,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09471","citing_title":"The Statistical Cost of Adaptation in Multi-Source Transfer Learning","ref_index":218,"is_internal_anchor":false},{"citing_arxiv_id":"2506.09985","citing_title":"V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU","json":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU.json","graph_json":"https://pith.science/api/pith-number/XL42NLYEUVGILAWPYCDW3VTGTU/graph.json","events_json":"https://pith.science/api/pith-number/XL42NLYEUVGILAWPYCDW3VTGTU/events.json","paper":"https://pith.science/paper/XL42NLYE"},"agent_actions":{"view_html":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU","download_json":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU.json","view_paper":"https://pith.science/paper/XL42NLYE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.07277&json=true","fetch_graph":"https://pith.science/api/pith-number/XL42NLYEUVGILAWPYCDW3VTGTU/graph.json","fetch_events":"https://pith.science/api/pith-number/XL42NLYEUVGILAWPYCDW3VTGTU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU/action/storage_attestation","attest_author":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU/action/author_attestation","sign_citation":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU/action/citation_signature","submit_replication":"https://pith.science/pith/XL42NLYEUVGILAWPYCDW3VTGTU/action/replication_record"}},"created_at":"2026-07-05T05:06:35.501650+00:00","updated_at":"2026-07-05T05:06:35.501650+00:00"}