{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:VT3OYF7GUAYL6MS67KMLI5QE73","short_pith_number":"pith:VT3OYF7G","schema_version":"1.0","canonical_sha256":"acf6ec17e6a030bf325efa98b47604fec0ea728ac91a3bdbee7df0dc70741921","source":{"kind":"arxiv","id":"2410.08326","version":1},"attestation_state":"computed","paper":{"title":"Neural Architecture Search of Hybrid Models for NPU-CIM Heterogeneous AR/VR Devices","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AR","cs.LG","cs.PF"],"primary_cat":"cs.CV","authors_text":"Barbara De Salvo, Chiao Liu, Jae-sun Seo, Jorge Tomas Gomez, Kleber Hugo Stangherlin, Phillip B. Gibbons, Sai Qian Zhang, Syed Shakib Sarwar, Win-San Khwa, Xiaoyu Sun, Yi-Lun Lu, Yiwei Zhao, Ziyun Li","submitted_at":"2024-10-10T19:30:34Z","abstract_excerpt":"Low-Latency and Low-Power Edge AI is essential for Virtual Reality and Augmented Reality applications. Recent advances show that hybrid models, combining convolution layers (CNN) and transformers (ViT), often achieve superior accuracy/performance tradeoff on various computer vision and machine learning (ML) tasks. However, hybrid ML models can pose system challenges for latency and energy-efficiency due to their diverse nature in dataflow and memory access patterns. In this work, we leverage the architecture heterogeneity from Neural Processing Units (NPU) and Compute-In-Memory (CIM) and perfo"},"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":"2410.08326","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-10T19:30:34Z","cross_cats_sorted":["cs.AR","cs.LG","cs.PF"],"title_canon_sha256":"ed24a4ce313919aa135bfc47a0ca807dc9d7a972c6e1b47d5097f46ff78c4d88","abstract_canon_sha256":"7c4fc14b8089ba10ba3f2ee6602bd7632350f7cda18cab501bc681d7069d1c71"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:56.385917Z","signature_b64":"m6yQxCB9Jsm/vC9Jlm2/K7t3YsxkBdd/Pw64vm3lQBWWsejZBbXRGHsWGrMYLVP9UgJTFk92kC3lK0LX7usYBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acf6ec17e6a030bf325efa98b47604fec0ea728ac91a3bdbee7df0dc70741921","last_reissued_at":"2026-07-05T09:18:56.385365Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:56.385365Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Architecture Search of Hybrid Models for NPU-CIM Heterogeneous AR/VR Devices","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AR","cs.LG","cs.PF"],"primary_cat":"cs.CV","authors_text":"Barbara De Salvo, Chiao Liu, Jae-sun Seo, Jorge Tomas Gomez, Kleber Hugo Stangherlin, Phillip B. Gibbons, Sai Qian Zhang, Syed Shakib Sarwar, Win-San Khwa, Xiaoyu Sun, Yi-Lun Lu, Yiwei Zhao, Ziyun Li","submitted_at":"2024-10-10T19:30:34Z","abstract_excerpt":"Low-Latency and Low-Power Edge AI is essential for Virtual Reality and Augmented Reality applications. Recent advances show that hybrid models, combining convolution layers (CNN) and transformers (ViT), often achieve superior accuracy/performance tradeoff on various computer vision and machine learning (ML) tasks. However, hybrid ML models can pose system challenges for latency and energy-efficiency due to their diverse nature in dataflow and memory access patterns. In this work, we leverage the architecture heterogeneity from Neural Processing Units (NPU) and Compute-In-Memory (CIM) and perfo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.08326","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/2410.08326/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":"2410.08326","created_at":"2026-07-05T09:18:56.385425+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.08326v1","created_at":"2026-07-05T09:18:56.385425+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.08326","created_at":"2026-07-05T09:18:56.385425+00:00"},{"alias_kind":"pith_short_12","alias_value":"VT3OYF7GUAYL","created_at":"2026-07-05T09:18:56.385425+00:00"},{"alias_kind":"pith_short_16","alias_value":"VT3OYF7GUAYL6MS6","created_at":"2026-07-05T09:18:56.385425+00:00"},{"alias_kind":"pith_short_8","alias_value":"VT3OYF7G","created_at":"2026-07-05T09:18:56.385425+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.04147","citing_title":"A3FR: Agile 3D Gaussian Splatting with Incremental Gaze Tracked Foveated Rendering in Virtual Reality","ref_index":82,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73","json":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73.json","graph_json":"https://pith.science/api/pith-number/VT3OYF7GUAYL6MS67KMLI5QE73/graph.json","events_json":"https://pith.science/api/pith-number/VT3OYF7GUAYL6MS67KMLI5QE73/events.json","paper":"https://pith.science/paper/VT3OYF7G"},"agent_actions":{"view_html":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73","download_json":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73.json","view_paper":"https://pith.science/paper/VT3OYF7G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.08326&json=true","fetch_graph":"https://pith.science/api/pith-number/VT3OYF7GUAYL6MS67KMLI5QE73/graph.json","fetch_events":"https://pith.science/api/pith-number/VT3OYF7GUAYL6MS67KMLI5QE73/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73/action/storage_attestation","attest_author":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73/action/author_attestation","sign_citation":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73/action/citation_signature","submit_replication":"https://pith.science/pith/VT3OYF7GUAYL6MS67KMLI5QE73/action/replication_record"}},"created_at":"2026-07-05T09:18:56.385425+00:00","updated_at":"2026-07-05T09:18:56.385425+00:00"}