{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:LTM5KF4MNJFSGN7B3D3VBTM4A7","short_pith_number":"pith:LTM5KF4M","schema_version":"1.0","canonical_sha256":"5cd9d5178c6a4b2337e1d8f750cd9c07f31999b8526b073182a0d99121f606a4","source":{"kind":"arxiv","id":"2205.11397","version":5},"attestation_state":"computed","paper":{"title":"Super Vision Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chunhua Shen, Liujuan Cao, Mengzhao Chen, Mingbao Lin, Rongrong Ji, Yuxin Zhang","submitted_at":"2022-05-23T15:42:12Z","abstract_excerpt":"We attempt to reduce the computational costs in vision transformers (ViTs), which increase quadratically in the token number. We present a novel training paradigm that trains only one ViT model at a time, but is capable of providing improved image recognition performance with various computational costs. Here, the trained ViT model, termed super vision transformer (SuperViT), is empowered with the versatile ability to solve incoming patches of multiple sizes as well as preserve informative tokens with multiple keeping rates (the ratio of keeping tokens) to achieve good hardware efficiency for "},"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":"2205.11397","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-05-23T15:42:12Z","cross_cats_sorted":[],"title_canon_sha256":"c9d0f6620d041a4ad00e0290ad21a80f442559667d6a60e0de2709b4ee14fd71","abstract_canon_sha256":"09f243a6d1ab95694621b5c8fe7b9e337ed729a5b2760992e0f5cd90bd2eee35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:32:33.311329Z","signature_b64":"wNWRodizQprsStJvwFjnepgNOxXDqalYoHE10az1IoJYGTlE709eibIasp7hS18f9pmnERgRSQpgvyIn7TmWDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cd9d5178c6a4b2337e1d8f750cd9c07f31999b8526b073182a0d99121f606a4","last_reissued_at":"2026-07-05T06:32:33.310818Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:32:33.310818Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Super Vision Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chunhua Shen, Liujuan Cao, Mengzhao Chen, Mingbao Lin, Rongrong Ji, Yuxin Zhang","submitted_at":"2022-05-23T15:42:12Z","abstract_excerpt":"We attempt to reduce the computational costs in vision transformers (ViTs), which increase quadratically in the token number. We present a novel training paradigm that trains only one ViT model at a time, but is capable of providing improved image recognition performance with various computational costs. Here, the trained ViT model, termed super vision transformer (SuperViT), is empowered with the versatile ability to solve incoming patches of multiple sizes as well as preserve informative tokens with multiple keeping rates (the ratio of keeping tokens) to achieve good hardware efficiency for "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.11397","kind":"arxiv","version":5},"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/2205.11397/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":"2205.11397","created_at":"2026-07-05T06:32:33.310884+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.11397v5","created_at":"2026-07-05T06:32:33.310884+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.11397","created_at":"2026-07-05T06:32:33.310884+00:00"},{"alias_kind":"pith_short_12","alias_value":"LTM5KF4MNJFS","created_at":"2026-07-05T06:32:33.310884+00:00"},{"alias_kind":"pith_short_16","alias_value":"LTM5KF4MNJFSGN7B","created_at":"2026-07-05T06:32:33.310884+00:00"},{"alias_kind":"pith_short_8","alias_value":"LTM5KF4M","created_at":"2026-07-05T06:32:33.310884+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/LTM5KF4MNJFSGN7B3D3VBTM4A7","json":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7.json","graph_json":"https://pith.science/api/pith-number/LTM5KF4MNJFSGN7B3D3VBTM4A7/graph.json","events_json":"https://pith.science/api/pith-number/LTM5KF4MNJFSGN7B3D3VBTM4A7/events.json","paper":"https://pith.science/paper/LTM5KF4M"},"agent_actions":{"view_html":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7","download_json":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7.json","view_paper":"https://pith.science/paper/LTM5KF4M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.11397&json=true","fetch_graph":"https://pith.science/api/pith-number/LTM5KF4MNJFSGN7B3D3VBTM4A7/graph.json","fetch_events":"https://pith.science/api/pith-number/LTM5KF4MNJFSGN7B3D3VBTM4A7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7/action/storage_attestation","attest_author":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7/action/author_attestation","sign_citation":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7/action/citation_signature","submit_replication":"https://pith.science/pith/LTM5KF4MNJFSGN7B3D3VBTM4A7/action/replication_record"}},"created_at":"2026-07-05T06:32:33.310884+00:00","updated_at":"2026-07-05T06:32:33.310884+00:00"}