{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SBLE6QW5OAHLLDNPZZESTWC3OT","short_pith_number":"pith:SBLE6QW5","schema_version":"1.0","canonical_sha256":"90564f42dd700eb58dafce4929d85b74f756698e87f5ce8704af055fccffa1cb","source":{"kind":"arxiv","id":"2203.04570","version":1},"attestation_state":"computed","paper":{"title":"CP-ViT: Cascade Vision Transformer Pruning via Progressive Sparsity Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Li Jiang, Naifeng Jing, Xiaoyao Liang, Yihong Xu, Zhezhi He, Zhuoran Song","submitted_at":"2022-03-09T08:15:14Z","abstract_excerpt":"Vision transformer (ViT) has achieved competitive accuracy on a variety of computer vision applications, but its computational cost impedes the deployment on resource-limited mobile devices.\n  We explore the sparsity in ViT and observe that informative patches and heads are sufficient for accurate image recognition.\n  In this paper, we propose a cascade pruning framework named CP-ViT by predicting sparsity in ViT models progressively and dynamically to reduce computational redundancy while minimizing the accuracy loss. Specifically, we define the cumulative score to reserve the informative pat"},"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":"2203.04570","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-03-09T08:15:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3d0c2258bee68d72cbf07dc9622759884b7fb952a62d8df7742763f9abe4c0e2","abstract_canon_sha256":"59bea122cbbfb47435f200026137425bb33a75c020a1bc6111b87ee790998dc4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:03:20.133655Z","signature_b64":"IYpgFm6oqSxaoHwZJn5Gme9+su97sqMG1n+OwuSpvxPf3H4sbqh40GL7AUwcJmrDAtRNWUzB9md7mtz8h5o0DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90564f42dd700eb58dafce4929d85b74f756698e87f5ce8704af055fccffa1cb","last_reissued_at":"2026-07-05T04:03:20.133216Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:03:20.133216Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CP-ViT: Cascade Vision Transformer Pruning via Progressive Sparsity Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Li Jiang, Naifeng Jing, Xiaoyao Liang, Yihong Xu, Zhezhi He, Zhuoran Song","submitted_at":"2022-03-09T08:15:14Z","abstract_excerpt":"Vision transformer (ViT) has achieved competitive accuracy on a variety of computer vision applications, but its computational cost impedes the deployment on resource-limited mobile devices.\n  We explore the sparsity in ViT and observe that informative patches and heads are sufficient for accurate image recognition.\n  In this paper, we propose a cascade pruning framework named CP-ViT by predicting sparsity in ViT models progressively and dynamically to reduce computational redundancy while minimizing the accuracy loss. Specifically, we define the cumulative score to reserve the informative pat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.04570","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/2203.04570/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":"2203.04570","created_at":"2026-07-05T04:03:20.133272+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.04570v1","created_at":"2026-07-05T04:03:20.133272+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.04570","created_at":"2026-07-05T04:03:20.133272+00:00"},{"alias_kind":"pith_short_12","alias_value":"SBLE6QW5OAHL","created_at":"2026-07-05T04:03:20.133272+00:00"},{"alias_kind":"pith_short_16","alias_value":"SBLE6QW5OAHLLDNP","created_at":"2026-07-05T04:03:20.133272+00:00"},{"alias_kind":"pith_short_8","alias_value":"SBLE6QW5","created_at":"2026-07-05T04:03:20.133272+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.18091","citing_title":"AutoSculpt: A Pattern-based Model Auto-pruning Framework Using Reinforcement Learning and Graph Learning","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17837","citing_title":"Temporal Aware Pruning for Efficient Diffusion-based Video Generation","ref_index":139,"is_internal_anchor":false},{"citing_arxiv_id":"2509.14165","citing_title":"Where Do Tokens Go? Understanding Pruning Behaviors in STEP at High Resolutions","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17837","citing_title":"Temporal Aware Pruning for Efficient Diffusion-based Video Generation","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2210.09461","citing_title":"Token Merging: Your ViT But Faster","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT","json":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT.json","graph_json":"https://pith.science/api/pith-number/SBLE6QW5OAHLLDNPZZESTWC3OT/graph.json","events_json":"https://pith.science/api/pith-number/SBLE6QW5OAHLLDNPZZESTWC3OT/events.json","paper":"https://pith.science/paper/SBLE6QW5"},"agent_actions":{"view_html":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT","download_json":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT.json","view_paper":"https://pith.science/paper/SBLE6QW5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.04570&json=true","fetch_graph":"https://pith.science/api/pith-number/SBLE6QW5OAHLLDNPZZESTWC3OT/graph.json","fetch_events":"https://pith.science/api/pith-number/SBLE6QW5OAHLLDNPZZESTWC3OT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT/action/storage_attestation","attest_author":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT/action/author_attestation","sign_citation":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT/action/citation_signature","submit_replication":"https://pith.science/pith/SBLE6QW5OAHLLDNPZZESTWC3OT/action/replication_record"}},"created_at":"2026-07-05T04:03:20.133272+00:00","updated_at":"2026-07-05T04:03:20.133272+00:00"}