{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:YUYWRT3BQSY5ILQKV6XDR5MBUU","short_pith_number":"pith:YUYWRT3B","schema_version":"1.0","canonical_sha256":"c53168cf6184b1d42e0aafae38f581a50c77704a371b0bfcbfa778c936e464eb","source":{"kind":"arxiv","id":"2405.10939","version":1},"attestation_state":"computed","paper":{"title":"DINO as a von Mises-Fisher mixture model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Fredrik Lindsten, Hariprasath Govindarajan, Jacob Roll, Per Sid\\'en","submitted_at":"2024-05-17T17:49:45Z","abstract_excerpt":"Self-distillation methods using Siamese networks are popular for self-supervised pre-training. DINO is one such method based on a cross-entropy loss between $K$-dimensional probability vectors, obtained by applying a softmax function to the dot product between representations and learnt prototypes. Given the fact that the learned representations are $L^2$-normalized, we show that DINO and its derivatives, such as iBOT, can be interpreted as a mixture model of von Mises-Fisher components. With this interpretation, DINO assumes equal precision for all components when the prototypes are also $L^2"},"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":"2405.10939","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-05-17T17:49:45Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"687f12da288bd54cbda462f722437f6b2b771ecd8bf1d9782b2666b9a68b8827","abstract_canon_sha256":"168637731ca5b392088306a7a1690c389864bf6a9ac2aca166743551866481c1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:20:19.636353Z","signature_b64":"BdXsVQZVQ9kyH5dhZPlbXGeORGCQlHKLLa9zK+1FoaGN1sSJuJ2tigZR8+vjd8Oz81D9RdzTgqZYueiJdxWyBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c53168cf6184b1d42e0aafae38f581a50c77704a371b0bfcbfa778c936e464eb","last_reissued_at":"2026-07-05T08:20:19.635873Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:20:19.635873Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DINO as a von Mises-Fisher mixture model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Fredrik Lindsten, Hariprasath Govindarajan, Jacob Roll, Per Sid\\'en","submitted_at":"2024-05-17T17:49:45Z","abstract_excerpt":"Self-distillation methods using Siamese networks are popular for self-supervised pre-training. DINO is one such method based on a cross-entropy loss between $K$-dimensional probability vectors, obtained by applying a softmax function to the dot product between representations and learnt prototypes. Given the fact that the learned representations are $L^2$-normalized, we show that DINO and its derivatives, such as iBOT, can be interpreted as a mixture model of von Mises-Fisher components. With this interpretation, DINO assumes equal precision for all components when the prototypes are also $L^2"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.10939","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/2405.10939/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":"2405.10939","created_at":"2026-07-05T08:20:19.635931+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.10939v1","created_at":"2026-07-05T08:20:19.635931+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.10939","created_at":"2026-07-05T08:20:19.635931+00:00"},{"alias_kind":"pith_short_12","alias_value":"YUYWRT3BQSY5","created_at":"2026-07-05T08:20:19.635931+00:00"},{"alias_kind":"pith_short_16","alias_value":"YUYWRT3BQSY5ILQK","created_at":"2026-07-05T08:20:19.635931+00:00"},{"alias_kind":"pith_short_8","alias_value":"YUYWRT3B","created_at":"2026-07-05T08:20:19.635931+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU","json":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU.json","graph_json":"https://pith.science/api/pith-number/YUYWRT3BQSY5ILQKV6XDR5MBUU/graph.json","events_json":"https://pith.science/api/pith-number/YUYWRT3BQSY5ILQKV6XDR5MBUU/events.json","paper":"https://pith.science/paper/YUYWRT3B"},"agent_actions":{"view_html":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU","download_json":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU.json","view_paper":"https://pith.science/paper/YUYWRT3B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.10939&json=true","fetch_graph":"https://pith.science/api/pith-number/YUYWRT3BQSY5ILQKV6XDR5MBUU/graph.json","fetch_events":"https://pith.science/api/pith-number/YUYWRT3BQSY5ILQKV6XDR5MBUU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU/action/storage_attestation","attest_author":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU/action/author_attestation","sign_citation":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU/action/citation_signature","submit_replication":"https://pith.science/pith/YUYWRT3BQSY5ILQKV6XDR5MBUU/action/replication_record"}},"created_at":"2026-07-05T08:20:19.635931+00:00","updated_at":"2026-07-05T08:20:19.635931+00:00"}