{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:JB2LBVHIUS43WZMU2QZ7A56TVJ","short_pith_number":"pith:JB2LBVHI","schema_version":"1.0","canonical_sha256":"4874b0d4e8a4b9bb6594d433f077d3aa7f9c5a8e15c1fab2e5d2ecaea32c8bcf","source":{"kind":"arxiv","id":"1911.04997","version":1},"attestation_state":"computed","paper":{"title":"Character-based NMT with Transformer","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Laurent Besacier, Marc Dymetman, Matthias Gall\\'e, Rohit Gupta","submitted_at":"2019-11-12T16:32:38Z","abstract_excerpt":"Character-based translation has several appealing advantages, but its performance is in general worse than a carefully tuned BPE baseline. In this paper we study the impact of character-based input and output with the Transformer architecture. In particular, our experiments on EN-DE show that character-based Transformer models are more robust than their BPE counterpart, both when translating noisy text, and when translating text from a different domain. To obtain comparable BLEU scores in clean, in-domain data and close the gap with BPE-based models we use known techniques to train deeper Tran"},"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":"1911.04997","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2019-11-12T16:32:38Z","cross_cats_sorted":[],"title_canon_sha256":"900d47a033502b199ec198e336be84106eda94de96440310406ff86e36ed6ebd","abstract_canon_sha256":"922b6a18ebf7826dc9609802d91ef8aa32437ca5107f53e94628c9d2775c81c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:18:46.916738Z","signature_b64":"/DxTTYgB/QWJxMNLRZXMMY40feUyLGEi/LZB/PRyrdccV/aSv1SQZzGSnvjz6oRHZD3HeImhO1SiCSs6k7UeDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4874b0d4e8a4b9bb6594d433f077d3aa7f9c5a8e15c1fab2e5d2ecaea32c8bcf","last_reissued_at":"2026-07-05T00:18:46.916359Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:18:46.916359Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Character-based NMT with Transformer","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Laurent Besacier, Marc Dymetman, Matthias Gall\\'e, Rohit Gupta","submitted_at":"2019-11-12T16:32:38Z","abstract_excerpt":"Character-based translation has several appealing advantages, but its performance is in general worse than a carefully tuned BPE baseline. In this paper we study the impact of character-based input and output with the Transformer architecture. In particular, our experiments on EN-DE show that character-based Transformer models are more robust than their BPE counterpart, both when translating noisy text, and when translating text from a different domain. To obtain comparable BLEU scores in clean, in-domain data and close the gap with BPE-based models we use known techniques to train deeper Tran"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.04997","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/1911.04997/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":"1911.04997","created_at":"2026-07-05T00:18:46.916418+00:00"},{"alias_kind":"arxiv_version","alias_value":"1911.04997v1","created_at":"2026-07-05T00:18:46.916418+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.04997","created_at":"2026-07-05T00:18:46.916418+00:00"},{"alias_kind":"pith_short_12","alias_value":"JB2LBVHIUS43","created_at":"2026-07-05T00:18:46.916418+00:00"},{"alias_kind":"pith_short_16","alias_value":"JB2LBVHIUS43WZMU","created_at":"2026-07-05T00:18:46.916418+00:00"},{"alias_kind":"pith_short_8","alias_value":"JB2LBVHI","created_at":"2026-07-05T00:18:46.916418+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.12417","citing_title":"HarMoEny: Efficient Multi-GPU Inference of MoE Models","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ","json":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ.json","graph_json":"https://pith.science/api/pith-number/JB2LBVHIUS43WZMU2QZ7A56TVJ/graph.json","events_json":"https://pith.science/api/pith-number/JB2LBVHIUS43WZMU2QZ7A56TVJ/events.json","paper":"https://pith.science/paper/JB2LBVHI"},"agent_actions":{"view_html":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ","download_json":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ.json","view_paper":"https://pith.science/paper/JB2LBVHI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1911.04997&json=true","fetch_graph":"https://pith.science/api/pith-number/JB2LBVHIUS43WZMU2QZ7A56TVJ/graph.json","fetch_events":"https://pith.science/api/pith-number/JB2LBVHIUS43WZMU2QZ7A56TVJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ/action/storage_attestation","attest_author":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ/action/author_attestation","sign_citation":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ/action/citation_signature","submit_replication":"https://pith.science/pith/JB2LBVHIUS43WZMU2QZ7A56TVJ/action/replication_record"}},"created_at":"2026-07-05T00:18:46.916418+00:00","updated_at":"2026-07-05T00:18:46.916418+00:00"}