{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:B5XYKMCZKT3RFCLAOXIALIQSID","short_pith_number":"pith:B5XYKMCZ","schema_version":"1.0","canonical_sha256":"0f6f85305954f712896075d005a21240c5150d341a18f38e1a1069c1b4f0b07d","source":{"kind":"arxiv","id":"1809.03985","version":1},"attestation_state":"computed","paper":{"title":"On The Alignment Problem In Multi-Head Attention-Based Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gabriel Bretschner, Hermann Ney, Tamer Alkhouli","submitted_at":"2018-09-11T15:41:12Z","abstract_excerpt":"This work investigates the alignment problem in state-of-the-art multi-head attention models based on the transformer architecture. We demonstrate that alignment extraction in transformer models can be improved by augmenting an additional alignment head to the multi-head source-to-target attention component. This is used to compute sharper attention weights. We describe how to use the alignment head to achieve competitive performance. To study the effect of adding the alignment head, we simulate a dictionary-guided translation task, where the user wants to guide translation using pre-defined d"},"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":"1809.03985","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-09-11T15:41:12Z","cross_cats_sorted":[],"title_canon_sha256":"31e5251744e1eaeff12f6103cf09983931785e2cb70c8a2ca35cfbbe6a19feba","abstract_canon_sha256":"c036b51f2be180f975fcdbf3f1a198a66b993b4488648daf46cc4876ce28bc6a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:05:57.644606Z","signature_b64":"T0ER72jsgH7mK2pUdQxmDdnOcr7i9H1WMqDoF7FoHbbgrJeTTLDB7xzrFsRzD1rjUfpM5IZ75dD8CwOKRRgCBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f6f85305954f712896075d005a21240c5150d341a18f38e1a1069c1b4f0b07d","last_reissued_at":"2026-05-18T00:05:57.644107Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:05:57.644107Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On The Alignment Problem In Multi-Head Attention-Based Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Gabriel Bretschner, Hermann Ney, Tamer Alkhouli","submitted_at":"2018-09-11T15:41:12Z","abstract_excerpt":"This work investigates the alignment problem in state-of-the-art multi-head attention models based on the transformer architecture. We demonstrate that alignment extraction in transformer models can be improved by augmenting an additional alignment head to the multi-head source-to-target attention component. This is used to compute sharper attention weights. We describe how to use the alignment head to achieve competitive performance. To study the effect of adding the alignment head, we simulate a dictionary-guided translation task, where the user wants to guide translation using pre-defined d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.03985","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":"1809.03985","created_at":"2026-05-18T00:05:57.644184+00:00"},{"alias_kind":"arxiv_version","alias_value":"1809.03985v1","created_at":"2026-05-18T00:05:57.644184+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.03985","created_at":"2026-05-18T00:05:57.644184+00:00"},{"alias_kind":"pith_short_12","alias_value":"B5XYKMCZKT3R","created_at":"2026-05-18T12:32:13.499390+00:00"},{"alias_kind":"pith_short_16","alias_value":"B5XYKMCZKT3RFCLA","created_at":"2026-05-18T12:32:13.499390+00:00"},{"alias_kind":"pith_short_8","alias_value":"B5XYKMCZ","created_at":"2026-05-18T12:32:13.499390+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/B5XYKMCZKT3RFCLAOXIALIQSID","json":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID.json","graph_json":"https://pith.science/api/pith-number/B5XYKMCZKT3RFCLAOXIALIQSID/graph.json","events_json":"https://pith.science/api/pith-number/B5XYKMCZKT3RFCLAOXIALIQSID/events.json","paper":"https://pith.science/paper/B5XYKMCZ"},"agent_actions":{"view_html":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID","download_json":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID.json","view_paper":"https://pith.science/paper/B5XYKMCZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1809.03985&json=true","fetch_graph":"https://pith.science/api/pith-number/B5XYKMCZKT3RFCLAOXIALIQSID/graph.json","fetch_events":"https://pith.science/api/pith-number/B5XYKMCZKT3RFCLAOXIALIQSID/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID/action/storage_attestation","attest_author":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID/action/author_attestation","sign_citation":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID/action/citation_signature","submit_replication":"https://pith.science/pith/B5XYKMCZKT3RFCLAOXIALIQSID/action/replication_record"}},"created_at":"2026-05-18T00:05:57.644184+00:00","updated_at":"2026-05-18T00:05:57.644184+00:00"}