{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:64LIEPLCEV2NSRK7JRXO4D6FSV","short_pith_number":"pith:64LIEPLC","schema_version":"1.0","canonical_sha256":"f716823d622574d9455f4c6eee0fc5956d97780a183f3a592044d3756963dcdb","source":{"kind":"arxiv","id":"2109.00642","version":1},"attestation_state":"computed","paper":{"title":"Searching for Efficient Multi-Stage Vision Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Sertac Karaman, Vivienne Sze, Yi-Lun Liao","submitted_at":"2021-09-01T22:37:56Z","abstract_excerpt":"Vision Transformer (ViT) demonstrates that Transformer for natural language processing can be applied to computer vision tasks and result in comparable performance to convolutional neural networks (CNN), which have been studied and adopted in computer vision for years. This naturally raises the question of how the performance of ViT can be advanced with design techniques of CNN. To this end, we propose to incorporate two techniques and present ViT-ResNAS, an efficient multi-stage ViT architecture designed with neural architecture search (NAS). First, we propose residual spatial reduction to de"},"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":"2109.00642","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-09-01T22:37:56Z","cross_cats_sorted":[],"title_canon_sha256":"4eb7f5a59749100127ae27f458ac265f0e555bb60a927d48979115f0e76e4b60","abstract_canon_sha256":"790a45889ab5d70fdbda926a4e85da818379776e1a047eade112874f225cedcf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:10:57.191384Z","signature_b64":"ePEWAkOnE4X2pudeytBdeT5Xbue6RKNAkLJW/XMbdrFSOyQ/COVbztok8WEblCmG4jPAxBmHU69yGDRi4qAlDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f716823d622574d9455f4c6eee0fc5956d97780a183f3a592044d3756963dcdb","last_reissued_at":"2026-07-05T03:10:57.190986Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:10:57.190986Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Searching for Efficient Multi-Stage Vision Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Sertac Karaman, Vivienne Sze, Yi-Lun Liao","submitted_at":"2021-09-01T22:37:56Z","abstract_excerpt":"Vision Transformer (ViT) demonstrates that Transformer for natural language processing can be applied to computer vision tasks and result in comparable performance to convolutional neural networks (CNN), which have been studied and adopted in computer vision for years. This naturally raises the question of how the performance of ViT can be advanced with design techniques of CNN. To this end, we propose to incorporate two techniques and present ViT-ResNAS, an efficient multi-stage ViT architecture designed with neural architecture search (NAS). First, we propose residual spatial reduction to de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.00642","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/2109.00642/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":"2109.00642","created_at":"2026-07-05T03:10:57.191048+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.00642v1","created_at":"2026-07-05T03:10:57.191048+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.00642","created_at":"2026-07-05T03:10:57.191048+00:00"},{"alias_kind":"pith_short_12","alias_value":"64LIEPLCEV2N","created_at":"2026-07-05T03:10:57.191048+00:00"},{"alias_kind":"pith_short_16","alias_value":"64LIEPLCEV2NSRK7","created_at":"2026-07-05T03:10:57.191048+00:00"},{"alias_kind":"pith_short_8","alias_value":"64LIEPLC","created_at":"2026-07-05T03:10:57.191048+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/64LIEPLCEV2NSRK7JRXO4D6FSV","json":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV.json","graph_json":"https://pith.science/api/pith-number/64LIEPLCEV2NSRK7JRXO4D6FSV/graph.json","events_json":"https://pith.science/api/pith-number/64LIEPLCEV2NSRK7JRXO4D6FSV/events.json","paper":"https://pith.science/paper/64LIEPLC"},"agent_actions":{"view_html":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV","download_json":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV.json","view_paper":"https://pith.science/paper/64LIEPLC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.00642&json=true","fetch_graph":"https://pith.science/api/pith-number/64LIEPLCEV2NSRK7JRXO4D6FSV/graph.json","fetch_events":"https://pith.science/api/pith-number/64LIEPLCEV2NSRK7JRXO4D6FSV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV/action/storage_attestation","attest_author":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV/action/author_attestation","sign_citation":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV/action/citation_signature","submit_replication":"https://pith.science/pith/64LIEPLCEV2NSRK7JRXO4D6FSV/action/replication_record"}},"created_at":"2026-07-05T03:10:57.191048+00:00","updated_at":"2026-07-05T03:10:57.191048+00:00"}