{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:DLECX7IOWSM7DPKCDJDTIVQQJH","short_pith_number":"pith:DLECX7IO","schema_version":"1.0","canonical_sha256":"1ac82bfd0eb499f1bd421a4734561049e7fc2a46a952609a309a7c5606c9afda","source":{"kind":"arxiv","id":"2305.07027","version":1},"attestation_state":"computed","paper":{"title":"EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Han Hu, Houwen Peng, Ningxin Zheng, Xinyu Liu, Yixuan Yuan, Yuqing Yang","submitted_at":"2023-05-11T17:59:41Z","abstract_excerpt":"Vision transformers have shown great success due to their high model capabilities. However, their remarkable performance is accompanied by heavy computation costs, which makes them unsuitable for real-time applications. In this paper, we propose a family of high-speed vision transformers named EfficientViT. We find that the speed of existing transformer models is commonly bounded by memory inefficient operations, especially the tensor reshaping and element-wise functions in MHSA. Therefore, we design a new building block with a sandwich layout, i.e., using a single memory-bound MHSA between ef"},"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":"2305.07027","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-11T17:59:41Z","cross_cats_sorted":[],"title_canon_sha256":"20abf59cfb3e2ec71db3e602a2d34570576787c717de6e67d3516d993efcc509","abstract_canon_sha256":"4a06f19901551ec3238ab2000b740468a2ebef940b81b641ce4ed032cf6c56b4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:09:18.737155Z","signature_b64":"4g9wwbJxj3tnYmG0rrss2j2Xzp+VZING5GgIwvjKDAX+XetIPAx5oEVwp4e4mER60yV/ceNod5owYrLjNAzYBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ac82bfd0eb499f1bd421a4734561049e7fc2a46a952609a309a7c5606c9afda","last_reissued_at":"2026-07-05T06:09:18.736649Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:09:18.736649Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Han Hu, Houwen Peng, Ningxin Zheng, Xinyu Liu, Yixuan Yuan, Yuqing Yang","submitted_at":"2023-05-11T17:59:41Z","abstract_excerpt":"Vision transformers have shown great success due to their high model capabilities. However, their remarkable performance is accompanied by heavy computation costs, which makes them unsuitable for real-time applications. In this paper, we propose a family of high-speed vision transformers named EfficientViT. We find that the speed of existing transformer models is commonly bounded by memory inefficient operations, especially the tensor reshaping and element-wise functions in MHSA. Therefore, we design a new building block with a sandwich layout, i.e., using a single memory-bound MHSA between ef"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.07027","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/2305.07027/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":"2305.07027","created_at":"2026-07-05T06:09:18.736711+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.07027v1","created_at":"2026-07-05T06:09:18.736711+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.07027","created_at":"2026-07-05T06:09:18.736711+00:00"},{"alias_kind":"pith_short_12","alias_value":"DLECX7IOWSM7","created_at":"2026-07-05T06:09:18.736711+00:00"},{"alias_kind":"pith_short_16","alias_value":"DLECX7IOWSM7DPKC","created_at":"2026-07-05T06:09:18.736711+00:00"},{"alias_kind":"pith_short_8","alias_value":"DLECX7IO","created_at":"2026-07-05T06:09:18.736711+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/DLECX7IOWSM7DPKCDJDTIVQQJH","json":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH.json","graph_json":"https://pith.science/api/pith-number/DLECX7IOWSM7DPKCDJDTIVQQJH/graph.json","events_json":"https://pith.science/api/pith-number/DLECX7IOWSM7DPKCDJDTIVQQJH/events.json","paper":"https://pith.science/paper/DLECX7IO"},"agent_actions":{"view_html":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH","download_json":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH.json","view_paper":"https://pith.science/paper/DLECX7IO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.07027&json=true","fetch_graph":"https://pith.science/api/pith-number/DLECX7IOWSM7DPKCDJDTIVQQJH/graph.json","fetch_events":"https://pith.science/api/pith-number/DLECX7IOWSM7DPKCDJDTIVQQJH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH/action/storage_attestation","attest_author":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH/action/author_attestation","sign_citation":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH/action/citation_signature","submit_replication":"https://pith.science/pith/DLECX7IOWSM7DPKCDJDTIVQQJH/action/replication_record"}},"created_at":"2026-07-05T06:09:18.736711+00:00","updated_at":"2026-07-05T06:09:18.736711+00:00"}