{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:HZV3RO4EBC2L7DHDPMSI7ZH6UF","short_pith_number":"pith:HZV3RO4E","schema_version":"1.0","canonical_sha256":"3e6bb8bb8408b4bf8ce37b248fe4fea15513fecd15a2144a712c64c4a92bdbc6","source":{"kind":"arxiv","id":"2201.12133","version":2},"attestation_state":"computed","paper":{"title":"O-ViT: Orthogonal Vision Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Mingsong Chen, Xian Wei, Yanhong Fei, Yingjie Liu","submitted_at":"2022-01-28T14:18:52Z","abstract_excerpt":"Inspired by the tremendous success of the self-attention mechanism in natural language processing, the Vision Transformer (ViT) creatively applies it to image patch sequences and achieves incredible performance. However, the scaled dot-product self-attention of ViT brings about scale ambiguity to the structure of the original feature space. To address this problem, we propose a novel method named Orthogonal Vision Transformer (O-ViT), to optimize ViT from the geometric perspective. O-ViT limits parameters of self-attention blocks to be on the norm-keeping orthogonal manifold, which can keep th"},"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":"2201.12133","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-01-28T14:18:52Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b8f21c4090b7d3899b15c9dab7739212b71b7720360e3098898dc76b2cb60e53","abstract_canon_sha256":"863c0d079fc22ea33be1672b56f461e37dd01b30fde65fd6e42a74605c024043"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:57:31.403972Z","signature_b64":"dJFH0iPqQNcGaVRZaZQHq982Wiour12HrK696p1RyRl5Wy9zfEr+Ne0FnQTFObdH3u+tTuAFO+6ketqC7hQLCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e6bb8bb8408b4bf8ce37b248fe4fea15513fecd15a2144a712c64c4a92bdbc6","last_reissued_at":"2026-07-05T03:57:31.403475Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:57:31.403475Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"O-ViT: Orthogonal Vision Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Mingsong Chen, Xian Wei, Yanhong Fei, Yingjie Liu","submitted_at":"2022-01-28T14:18:52Z","abstract_excerpt":"Inspired by the tremendous success of the self-attention mechanism in natural language processing, the Vision Transformer (ViT) creatively applies it to image patch sequences and achieves incredible performance. However, the scaled dot-product self-attention of ViT brings about scale ambiguity to the structure of the original feature space. To address this problem, we propose a novel method named Orthogonal Vision Transformer (O-ViT), to optimize ViT from the geometric perspective. O-ViT limits parameters of self-attention blocks to be on the norm-keeping orthogonal manifold, which can keep th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.12133","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/2201.12133/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":"2201.12133","created_at":"2026-07-05T03:57:31.403535+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.12133v2","created_at":"2026-07-05T03:57:31.403535+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.12133","created_at":"2026-07-05T03:57:31.403535+00:00"},{"alias_kind":"pith_short_12","alias_value":"HZV3RO4EBC2L","created_at":"2026-07-05T03:57:31.403535+00:00"},{"alias_kind":"pith_short_16","alias_value":"HZV3RO4EBC2L7DHD","created_at":"2026-07-05T03:57:31.403535+00:00"},{"alias_kind":"pith_short_8","alias_value":"HZV3RO4E","created_at":"2026-07-05T03:57:31.403535+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2601.00417","citing_title":"Deep Delta Learning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04087","citing_title":"BOOOM: Loss-Function-Agnostic Black-Box Optimization over Orthonormal Manifolds for Machine Learning and Statistical Inference","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF","json":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF.json","graph_json":"https://pith.science/api/pith-number/HZV3RO4EBC2L7DHDPMSI7ZH6UF/graph.json","events_json":"https://pith.science/api/pith-number/HZV3RO4EBC2L7DHDPMSI7ZH6UF/events.json","paper":"https://pith.science/paper/HZV3RO4E"},"agent_actions":{"view_html":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF","download_json":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF.json","view_paper":"https://pith.science/paper/HZV3RO4E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.12133&json=true","fetch_graph":"https://pith.science/api/pith-number/HZV3RO4EBC2L7DHDPMSI7ZH6UF/graph.json","fetch_events":"https://pith.science/api/pith-number/HZV3RO4EBC2L7DHDPMSI7ZH6UF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF/action/storage_attestation","attest_author":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF/action/author_attestation","sign_citation":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF/action/citation_signature","submit_replication":"https://pith.science/pith/HZV3RO4EBC2L7DHDPMSI7ZH6UF/action/replication_record"}},"created_at":"2026-07-05T03:57:31.403535+00:00","updated_at":"2026-07-05T03:57:31.403535+00:00"}