{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:BYV3WPKHURVO5LUGVNUXLH5APU","short_pith_number":"pith:BYV3WPKH","schema_version":"1.0","canonical_sha256":"0e2bbb3d47a46aeeae86ab69759fa07d25a89335c790ead07d831e6d6e288440","source":{"kind":"arxiv","id":"2107.08199","version":2},"attestation_state":"computed","paper":{"title":"Dynamic Transformer for Efficient Machine Translation on Embedded Devices","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Amin Sabet, Geoff V. Merrett, Hishan Parry, Jia Bi, Jonathon Hare, Lei Xun","submitted_at":"2021-07-17T07:36:29Z","abstract_excerpt":"The Transformer architecture is widely used for machine translation tasks. However, its resource-intensive nature makes it challenging to implement on constrained embedded devices, particularly where available hardware resources can vary at run-time. We propose a dynamic machine translation model that scales the Transformer architecture based on the available resources at any particular time. The proposed approach, 'Dynamic-HAT', uses a HAT SuperTransformer as the backbone to search for SubTransformers with different accuracy-latency trade-offs at design time. The optimal SubTransformers are s"},"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":"2107.08199","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-07-17T07:36:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"280bdb9403e4c86b22c07f10350601017616e701e4d72ecff00ae2a1cf01f616","abstract_canon_sha256":"70747a0f6c73de414d337086dcf90d0eb3995c785540ebb682f1898a15d0e17d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:02:17.198361Z","signature_b64":"GLaJKnkl7r1osIDrSCmhDH8RFwst/upQ7eiUPPJwH4Q9FDYtK5w2UZuD7rHM8++oIhEazh2JRZwLlRLas3bpBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0e2bbb3d47a46aeeae86ab69759fa07d25a89335c790ead07d831e6d6e288440","last_reissued_at":"2026-07-05T03:02:17.197984Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:02:17.197984Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dynamic Transformer for Efficient Machine Translation on Embedded Devices","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Amin Sabet, Geoff V. Merrett, Hishan Parry, Jia Bi, Jonathon Hare, Lei Xun","submitted_at":"2021-07-17T07:36:29Z","abstract_excerpt":"The Transformer architecture is widely used for machine translation tasks. However, its resource-intensive nature makes it challenging to implement on constrained embedded devices, particularly where available hardware resources can vary at run-time. We propose a dynamic machine translation model that scales the Transformer architecture based on the available resources at any particular time. The proposed approach, 'Dynamic-HAT', uses a HAT SuperTransformer as the backbone to search for SubTransformers with different accuracy-latency trade-offs at design time. The optimal SubTransformers are s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.08199","kind":"arxiv","version":2},"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/2107.08199/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":"2107.08199","created_at":"2026-07-05T03:02:17.198040+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.08199v2","created_at":"2026-07-05T03:02:17.198040+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.08199","created_at":"2026-07-05T03:02:17.198040+00:00"},{"alias_kind":"pith_short_12","alias_value":"BYV3WPKHURVO","created_at":"2026-07-05T03:02:17.198040+00:00"},{"alias_kind":"pith_short_16","alias_value":"BYV3WPKHURVO5LUG","created_at":"2026-07-05T03:02:17.198040+00:00"},{"alias_kind":"pith_short_8","alias_value":"BYV3WPKH","created_at":"2026-07-05T03:02:17.198040+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/BYV3WPKHURVO5LUGVNUXLH5APU","json":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU.json","graph_json":"https://pith.science/api/pith-number/BYV3WPKHURVO5LUGVNUXLH5APU/graph.json","events_json":"https://pith.science/api/pith-number/BYV3WPKHURVO5LUGVNUXLH5APU/events.json","paper":"https://pith.science/paper/BYV3WPKH"},"agent_actions":{"view_html":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU","download_json":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU.json","view_paper":"https://pith.science/paper/BYV3WPKH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.08199&json=true","fetch_graph":"https://pith.science/api/pith-number/BYV3WPKHURVO5LUGVNUXLH5APU/graph.json","fetch_events":"https://pith.science/api/pith-number/BYV3WPKHURVO5LUGVNUXLH5APU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU/action/storage_attestation","attest_author":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU/action/author_attestation","sign_citation":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU/action/citation_signature","submit_replication":"https://pith.science/pith/BYV3WPKHURVO5LUGVNUXLH5APU/action/replication_record"}},"created_at":"2026-07-05T03:02:17.198040+00:00","updated_at":"2026-07-05T03:02:17.198040+00:00"}