{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ENDOT2YL7RE6LDQSRYJWJA6MDY","short_pith_number":"pith:ENDOT2YL","schema_version":"1.0","canonical_sha256":"2346e9eb0bfc49e58e128e136483cc1e34426a5db80d78b8c6937c7c12aa3c42","source":{"kind":"arxiv","id":"2606.19913","version":1},"attestation_state":"computed","paper":{"title":"Design and Evaluation of Energy-Efficient Whisper Dot-Product Kernel Offloading on a CGLA Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Ayumu Takeuchi, Takuto Ando, Yasuhiko Nakashima, Yu Eto","submitted_at":"2026-06-18T08:07:59Z","abstract_excerpt":"In this paper, we implement and evaluate Whisper dot-product kernel offloading on IMAX, a programmable Coarse-Grained Linear Arrays (CGLAs) architecture. Whisper-tiny.en profiling on an ARM Cortex-A72 shows that dot-product operations account for 90.6% of FP16 execution time and 87.1% of Q8_0 execution time. To address this kernel bottleneck, we combine kernel mapping, local-memory sizing, and burst scheduling. The implementation uses inline FP16-to-FP32 conversion, 2-way SIMD FMA on a 64-bit datapath, column-wise multithreading, and mixed execution in which aligned vector segments run on IMAX"},"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":"2606.19913","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2026-06-18T08:07:59Z","cross_cats_sorted":[],"title_canon_sha256":"3d928ba60d1fe33723d0bdce8ac1fe9175b951baa656996cf03978eb9de97eba","abstract_canon_sha256":"11c6c3931e5d62d4e7249205e5ae24a0897ae4e6c748b3750e1d0829e57f4b37"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:38.524193Z","signature_b64":"Oie/0VqEWRxYy6F7Q/dg9cXpi+DWihvyZWadEjx5SKjx508IwZYvYRMG7WcfWBdJfOA7G07fgeywGMQJengXDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2346e9eb0bfc49e58e128e136483cc1e34426a5db80d78b8c6937c7c12aa3c42","last_reissued_at":"2026-06-19T16:12:38.523795Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:38.523795Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Design and Evaluation of Energy-Efficient Whisper Dot-Product Kernel Offloading on a CGLA Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Ayumu Takeuchi, Takuto Ando, Yasuhiko Nakashima, Yu Eto","submitted_at":"2026-06-18T08:07:59Z","abstract_excerpt":"In this paper, we implement and evaluate Whisper dot-product kernel offloading on IMAX, a programmable Coarse-Grained Linear Arrays (CGLAs) architecture. Whisper-tiny.en profiling on an ARM Cortex-A72 shows that dot-product operations account for 90.6% of FP16 execution time and 87.1% of Q8_0 execution time. To address this kernel bottleneck, we combine kernel mapping, local-memory sizing, and burst scheduling. The implementation uses inline FP16-to-FP32 conversion, 2-way SIMD FMA on a 64-bit datapath, column-wise multithreading, and mixed execution in which aligned vector segments run on IMAX"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19913","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/2606.19913/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":"2606.19913","created_at":"2026-06-19T16:12:38.523859+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.19913v1","created_at":"2026-06-19T16:12:38.523859+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19913","created_at":"2026-06-19T16:12:38.523859+00:00"},{"alias_kind":"pith_short_12","alias_value":"ENDOT2YL7RE6","created_at":"2026-06-19T16:12:38.523859+00:00"},{"alias_kind":"pith_short_16","alias_value":"ENDOT2YL7RE6LDQS","created_at":"2026-06-19T16:12:38.523859+00:00"},{"alias_kind":"pith_short_8","alias_value":"ENDOT2YL","created_at":"2026-06-19T16:12:38.523859+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/ENDOT2YL7RE6LDQSRYJWJA6MDY","json":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY.json","graph_json":"https://pith.science/api/pith-number/ENDOT2YL7RE6LDQSRYJWJA6MDY/graph.json","events_json":"https://pith.science/api/pith-number/ENDOT2YL7RE6LDQSRYJWJA6MDY/events.json","paper":"https://pith.science/paper/ENDOT2YL"},"agent_actions":{"view_html":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY","download_json":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY.json","view_paper":"https://pith.science/paper/ENDOT2YL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.19913&json=true","fetch_graph":"https://pith.science/api/pith-number/ENDOT2YL7RE6LDQSRYJWJA6MDY/graph.json","fetch_events":"https://pith.science/api/pith-number/ENDOT2YL7RE6LDQSRYJWJA6MDY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY/action/storage_attestation","attest_author":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY/action/author_attestation","sign_citation":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY/action/citation_signature","submit_replication":"https://pith.science/pith/ENDOT2YL7RE6LDQSRYJWJA6MDY/action/replication_record"}},"created_at":"2026-06-19T16:12:38.523859+00:00","updated_at":"2026-06-19T16:12:38.523859+00:00"}