{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PU42ZCGM4K3TSIJMVECDE5IYDA","short_pith_number":"pith:PU42ZCGM","schema_version":"1.0","canonical_sha256":"7d39ac88cce2b739212ca904327518180561c31b99f32b1635c5320ca719de84","source":{"kind":"arxiv","id":"2412.16491","version":1},"attestation_state":"computed","paper":{"title":"ImagePiece: Content-aware Re-tokenization for Efficient Image Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bumsoo Kim, Hyeseung Cho, Seungdong Yoa, Seungjun Lee, Woohyung Lim","submitted_at":"2024-12-21T05:38:20Z","abstract_excerpt":"Vision Transformers (ViTs) have achieved remarkable success in various computer vision tasks. However, ViTs have a huge computational cost due to their inherent reliance on multi-head self-attention (MHSA), prompting efforts to accelerate ViTs for practical applications. To this end, recent works aim to reduce the number of tokens, mainly focusing on how to effectively prune or merge them. Nevertheless, since ViT tokens are generated from non-overlapping grid patches, they usually do not convey sufficient semantics, making it incompatible with efficient ViTs. To address this, we propose ImageP"},"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":"2412.16491","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-21T05:38:20Z","cross_cats_sorted":[],"title_canon_sha256":"a2964c391c83af11adbbb40aa1cff0cb2ea95e7f172508982e75669c331770f4","abstract_canon_sha256":"13da4d0e0bfbd18a9b53ba26b76f29c6046ea3fb1be686063ba7cfec60ef7471"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:48.661960Z","signature_b64":"h8scN/Y8on5Lz1ps8pA0dfqdArAcxbeOscfEY5UJ6n9cNBvaKHt6xRGFDxz1mH9L1VufEF+bY/nNgamnaXMPCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d39ac88cce2b739212ca904327518180561c31b99f32b1635c5320ca719de84","last_reissued_at":"2026-07-05T09:52:48.661489Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:48.661489Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ImagePiece: Content-aware Re-tokenization for Efficient Image Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bumsoo Kim, Hyeseung Cho, Seungdong Yoa, Seungjun Lee, Woohyung Lim","submitted_at":"2024-12-21T05:38:20Z","abstract_excerpt":"Vision Transformers (ViTs) have achieved remarkable success in various computer vision tasks. However, ViTs have a huge computational cost due to their inherent reliance on multi-head self-attention (MHSA), prompting efforts to accelerate ViTs for practical applications. To this end, recent works aim to reduce the number of tokens, mainly focusing on how to effectively prune or merge them. Nevertheless, since ViT tokens are generated from non-overlapping grid patches, they usually do not convey sufficient semantics, making it incompatible with efficient ViTs. To address this, we propose ImageP"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.16491","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/2412.16491/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":"2412.16491","created_at":"2026-07-05T09:52:48.661552+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.16491v1","created_at":"2026-07-05T09:52:48.661552+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.16491","created_at":"2026-07-05T09:52:48.661552+00:00"},{"alias_kind":"pith_short_12","alias_value":"PU42ZCGM4K3T","created_at":"2026-07-05T09:52:48.661552+00:00"},{"alias_kind":"pith_short_16","alias_value":"PU42ZCGM4K3TSIJM","created_at":"2026-07-05T09:52:48.661552+00:00"},{"alias_kind":"pith_short_8","alias_value":"PU42ZCGM","created_at":"2026-07-05T09:52:48.661552+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/PU42ZCGM4K3TSIJMVECDE5IYDA","json":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA.json","graph_json":"https://pith.science/api/pith-number/PU42ZCGM4K3TSIJMVECDE5IYDA/graph.json","events_json":"https://pith.science/api/pith-number/PU42ZCGM4K3TSIJMVECDE5IYDA/events.json","paper":"https://pith.science/paper/PU42ZCGM"},"agent_actions":{"view_html":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA","download_json":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA.json","view_paper":"https://pith.science/paper/PU42ZCGM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.16491&json=true","fetch_graph":"https://pith.science/api/pith-number/PU42ZCGM4K3TSIJMVECDE5IYDA/graph.json","fetch_events":"https://pith.science/api/pith-number/PU42ZCGM4K3TSIJMVECDE5IYDA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA/action/storage_attestation","attest_author":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA/action/author_attestation","sign_citation":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA/action/citation_signature","submit_replication":"https://pith.science/pith/PU42ZCGM4K3TSIJMVECDE5IYDA/action/replication_record"}},"created_at":"2026-07-05T09:52:48.661552+00:00","updated_at":"2026-07-05T09:52:48.661552+00:00"}