{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LYZZJV3A5A4B64YLWEC5RCLJA4","short_pith_number":"pith:LYZZJV3A","schema_version":"1.0","canonical_sha256":"5e3394d760e8381f730bb105d8896907361a8a953b68f0b9ff7f6e3f8c08858d","source":{"kind":"arxiv","id":"2606.11257","version":1},"attestation_state":"computed","paper":{"title":"Energy-Efficient On-Device RAG on a Mobile NPU: System Design and Benchmark on Snapdragon X Elite","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.PF"],"primary_cat":"cs.CL","authors_text":"Longying Lai, Zhiyuan Cheng","submitted_at":"2026-06-09T01:09:00Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) pipelines are compute-intensive, combining embedding, retrieval, reranking, and large language model (LLM) generation. Running them entirely on-device benefits privacy, latency, and offline use, but the energy cost of CPU inference is a major barrier. We present what is, to our knowledge, the first end-to-end RAG pipeline that runs all neural stages -- embedding, reranking, and LLM generation -- on the Qualcomm Hexagon NPU of the Snapdragon X Elite. Profiling on a Dell XPS 13 laptop, we compare NPU-accelerated RAG against CPU and OpenCL/Adreno GPU baselines"},"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":"2606.11257","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-09T01:09:00Z","cross_cats_sorted":["cs.LG","cs.PF"],"title_canon_sha256":"58eca095bd76d0070b4fb13fd41538f61f5e2dd32f68e221771040074a19a6d1","abstract_canon_sha256":"fd1c0446c9e8f6d32a42ec3dd47a547fe680f075c769256119e5f633642c9614"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T00:08:15.092961Z","signature_b64":"G23BFgoVEK5GW0dnPYas+/6eziQhpciPZbZd7JnGKiuzM58G9hWyIboM0i0PqVVZMgJaU8rBFEgQv1GSSFxGCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5e3394d760e8381f730bb105d8896907361a8a953b68f0b9ff7f6e3f8c08858d","last_reissued_at":"2026-06-11T00:08:15.092135Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T00:08:15.092135Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Energy-Efficient On-Device RAG on a Mobile NPU: System Design and Benchmark on Snapdragon X Elite","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.PF"],"primary_cat":"cs.CL","authors_text":"Longying Lai, Zhiyuan Cheng","submitted_at":"2026-06-09T01:09:00Z","abstract_excerpt":"Retrieval-Augmented Generation (RAG) pipelines are compute-intensive, combining embedding, retrieval, reranking, and large language model (LLM) generation. Running them entirely on-device benefits privacy, latency, and offline use, but the energy cost of CPU inference is a major barrier. We present what is, to our knowledge, the first end-to-end RAG pipeline that runs all neural stages -- embedding, reranking, and LLM generation -- on the Qualcomm Hexagon NPU of the Snapdragon X Elite. Profiling on a Dell XPS 13 laptop, we compare NPU-accelerated RAG against CPU and OpenCL/Adreno GPU baselines"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11257","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/2606.11257/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":"2606.11257","created_at":"2026-06-11T00:08:15.092270+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.11257v1","created_at":"2026-06-11T00:08:15.092270+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11257","created_at":"2026-06-11T00:08:15.092270+00:00"},{"alias_kind":"pith_short_12","alias_value":"LYZZJV3A5A4B","created_at":"2026-06-11T00:08:15.092270+00:00"},{"alias_kind":"pith_short_16","alias_value":"LYZZJV3A5A4B64YL","created_at":"2026-06-11T00:08:15.092270+00:00"},{"alias_kind":"pith_short_8","alias_value":"LYZZJV3A","created_at":"2026-06-11T00:08:15.092270+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/LYZZJV3A5A4B64YLWEC5RCLJA4","json":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4.json","graph_json":"https://pith.science/api/pith-number/LYZZJV3A5A4B64YLWEC5RCLJA4/graph.json","events_json":"https://pith.science/api/pith-number/LYZZJV3A5A4B64YLWEC5RCLJA4/events.json","paper":"https://pith.science/paper/LYZZJV3A"},"agent_actions":{"view_html":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4","download_json":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4.json","view_paper":"https://pith.science/paper/LYZZJV3A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.11257&json=true","fetch_graph":"https://pith.science/api/pith-number/LYZZJV3A5A4B64YLWEC5RCLJA4/graph.json","fetch_events":"https://pith.science/api/pith-number/LYZZJV3A5A4B64YLWEC5RCLJA4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4/action/storage_attestation","attest_author":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4/action/author_attestation","sign_citation":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4/action/citation_signature","submit_replication":"https://pith.science/pith/LYZZJV3A5A4B64YLWEC5RCLJA4/action/replication_record"}},"created_at":"2026-06-11T00:08:15.092270+00:00","updated_at":"2026-06-11T00:08:15.092270+00:00"}