{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:6S67FQUCS3M366JL5FHCVHH6TB","short_pith_number":"pith:6S67FQUC","schema_version":"1.0","canonical_sha256":"f4bdf2c28296d9bf792be94e2a9cfe984cb56f3e0fe057da64d66c554e655f0b","source":{"kind":"arxiv","id":"2203.10833","version":2},"attestation_state":"computed","paper":{"title":"Hyperbolic Vision Transformers: Combining Improvements in Metric Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aleksandr Ermolov, Ivan Oseledets, Leyla Mirvakhabova, Nicu Sebe, Valentin Khrulkov","submitted_at":"2022-03-21T09:48:23Z","abstract_excerpt":"Metric learning aims to learn a highly discriminative model encouraging the embeddings of similar classes to be close in the chosen metrics and pushed apart for dissimilar ones. The common recipe is to use an encoder to extract embeddings and a distance-based loss function to match the representations -- usually, the Euclidean distance is utilized. An emerging interest in learning hyperbolic data embeddings suggests that hyperbolic geometry can be beneficial for natural data. Following this line of work, we propose a new hyperbolic-based model for metric learning. At the core of our method is "},"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":"2203.10833","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-03-21T09:48:23Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a0bfa75595676e0b1bb66809414956ac547e1cd53973455cb6bb2198349dde2f","abstract_canon_sha256":"b945482f48c78dee58e19b622ab91e8aa500693c0933335797ea47a9d928e9a8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:07:09.304804Z","signature_b64":"Nc0GRRov5i07Htlxv8alzlAxj3Zg6NKJw2SuIezvx7UnnfMeHTCST6U0Mw3frF1frpRN0f/HQkaCj/wbomuWAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f4bdf2c28296d9bf792be94e2a9cfe984cb56f3e0fe057da64d66c554e655f0b","last_reissued_at":"2026-07-05T04:07:09.304405Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:07:09.304405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hyperbolic Vision Transformers: Combining Improvements in Metric Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Aleksandr Ermolov, Ivan Oseledets, Leyla Mirvakhabova, Nicu Sebe, Valentin Khrulkov","submitted_at":"2022-03-21T09:48:23Z","abstract_excerpt":"Metric learning aims to learn a highly discriminative model encouraging the embeddings of similar classes to be close in the chosen metrics and pushed apart for dissimilar ones. The common recipe is to use an encoder to extract embeddings and a distance-based loss function to match the representations -- usually, the Euclidean distance is utilized. An emerging interest in learning hyperbolic data embeddings suggests that hyperbolic geometry can be beneficial for natural data. Following this line of work, we propose a new hyperbolic-based model for metric learning. At the core of our method is "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.10833","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/2203.10833/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":"2203.10833","created_at":"2026-07-05T04:07:09.304464+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.10833v2","created_at":"2026-07-05T04:07:09.304464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.10833","created_at":"2026-07-05T04:07:09.304464+00:00"},{"alias_kind":"pith_short_12","alias_value":"6S67FQUCS3M3","created_at":"2026-07-05T04:07:09.304464+00:00"},{"alias_kind":"pith_short_16","alias_value":"6S67FQUCS3M366JL","created_at":"2026-07-05T04:07:09.304464+00:00"},{"alias_kind":"pith_short_8","alias_value":"6S67FQUC","created_at":"2026-07-05T04:07:09.304464+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.21027","citing_title":"HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering","ref_index":226,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB","json":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB.json","graph_json":"https://pith.science/api/pith-number/6S67FQUCS3M366JL5FHCVHH6TB/graph.json","events_json":"https://pith.science/api/pith-number/6S67FQUCS3M366JL5FHCVHH6TB/events.json","paper":"https://pith.science/paper/6S67FQUC"},"agent_actions":{"view_html":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB","download_json":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB.json","view_paper":"https://pith.science/paper/6S67FQUC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.10833&json=true","fetch_graph":"https://pith.science/api/pith-number/6S67FQUCS3M366JL5FHCVHH6TB/graph.json","fetch_events":"https://pith.science/api/pith-number/6S67FQUCS3M366JL5FHCVHH6TB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB/action/storage_attestation","attest_author":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB/action/author_attestation","sign_citation":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB/action/citation_signature","submit_replication":"https://pith.science/pith/6S67FQUCS3M366JL5FHCVHH6TB/action/replication_record"}},"created_at":"2026-07-05T04:07:09.304464+00:00","updated_at":"2026-07-05T04:07:09.304464+00:00"}