{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:HLNAL6DHLSI3WDAKK6JQKARV74","short_pith_number":"pith:HLNAL6DH","schema_version":"1.0","canonical_sha256":"3ada05f8675c91bb0c0a5793050235ff0bda0ae3ded6ce016166bba144e6df49","source":{"kind":"arxiv","id":"1906.11024","version":1},"attestation_state":"computed","paper":{"title":"Sharing Attention Weights for Fast Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jingbo Zhu, Tongran Liu, Tong Xiao, Yinqiao Li, Zhengtao Yu","submitted_at":"2019-06-26T12:27:05Z","abstract_excerpt":"Recently, the Transformer machine translation system has shown strong results by stacking attention layers on both the source and target-language sides. But the inference of this model is slow due to the heavy use of dot-product attention in auto-regressive decoding. In this paper we speed up Transformer via a fast and lightweight attention model. More specifically, we share attention weights in adjacent layers and enable the efficient re-use of hidden states in a vertical manner. Moreover, the sharing policy can be jointly learned with the MT model. We test our approach on ten WMT and NIST Op"},"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":"1906.11024","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-06-26T12:27:05Z","cross_cats_sorted":[],"title_canon_sha256":"de72a14375ba101939e98add75c4980cb6f87cb780e229741bda14dbe5d22a74","abstract_canon_sha256":"f2ab3541e8af71ed4f5218d4e35d2f5407063b748ad7ce58bee349c2dca7bae4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:10.181002Z","signature_b64":"SV8m3M+aZE7qUNW1uj7OmvE0FlnkF5sSUJiKVBCublj2eBmGPoPoE8uzm5dCot0jtHqnv9rxHnjI2pFjWL5GCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ada05f8675c91bb0c0a5793050235ff0bda0ae3ded6ce016166bba144e6df49","last_reissued_at":"2026-05-17T23:42:10.180301Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:10.180301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sharing Attention Weights for Fast Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jingbo Zhu, Tongran Liu, Tong Xiao, Yinqiao Li, Zhengtao Yu","submitted_at":"2019-06-26T12:27:05Z","abstract_excerpt":"Recently, the Transformer machine translation system has shown strong results by stacking attention layers on both the source and target-language sides. But the inference of this model is slow due to the heavy use of dot-product attention in auto-regressive decoding. In this paper we speed up Transformer via a fast and lightweight attention model. More specifically, we share attention weights in adjacent layers and enable the efficient re-use of hidden states in a vertical manner. Moreover, the sharing policy can be jointly learned with the MT model. We test our approach on ten WMT and NIST Op"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.11024","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":""},"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":"1906.11024","created_at":"2026-05-17T23:42:10.180424+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.11024v1","created_at":"2026-05-17T23:42:10.180424+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.11024","created_at":"2026-05-17T23:42:10.180424+00:00"},{"alias_kind":"pith_short_12","alias_value":"HLNAL6DHLSI3","created_at":"2026-05-18T12:33:18.533446+00:00"},{"alias_kind":"pith_short_16","alias_value":"HLNAL6DHLSI3WDAK","created_at":"2026-05-18T12:33:18.533446+00:00"},{"alias_kind":"pith_short_8","alias_value":"HLNAL6DH","created_at":"2026-05-18T12:33:18.533446+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/HLNAL6DHLSI3WDAKK6JQKARV74","json":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74.json","graph_json":"https://pith.science/api/pith-number/HLNAL6DHLSI3WDAKK6JQKARV74/graph.json","events_json":"https://pith.science/api/pith-number/HLNAL6DHLSI3WDAKK6JQKARV74/events.json","paper":"https://pith.science/paper/HLNAL6DH"},"agent_actions":{"view_html":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74","download_json":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74.json","view_paper":"https://pith.science/paper/HLNAL6DH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.11024&json=true","fetch_graph":"https://pith.science/api/pith-number/HLNAL6DHLSI3WDAKK6JQKARV74/graph.json","fetch_events":"https://pith.science/api/pith-number/HLNAL6DHLSI3WDAKK6JQKARV74/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74/action/storage_attestation","attest_author":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74/action/author_attestation","sign_citation":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74/action/citation_signature","submit_replication":"https://pith.science/pith/HLNAL6DHLSI3WDAKK6JQKARV74/action/replication_record"}},"created_at":"2026-05-17T23:42:10.180424+00:00","updated_at":"2026-05-17T23:42:10.180424+00:00"}