{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:P4LN274PCH4QIX2DKME7GGSQOB","short_pith_number":"pith:P4LN274P","schema_version":"1.0","canonical_sha256":"7f16dd7f8f11f9045f435309f31a507048a0d222486639fc5a3e96e698635449","source":{"kind":"arxiv","id":"2305.14189","version":3},"attestation_state":"computed","paper":{"title":"Beyond Shared Vocabulary: Increasing Representational Word Similarities across Languages for Multilingual Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christof Monz, Di Wu","submitted_at":"2023-05-23T16:11:00Z","abstract_excerpt":"Using a vocabulary that is shared across languages is common practice in Multilingual Neural Machine Translation (MNMT). In addition to its simple design, shared tokens play an important role in positive knowledge transfer, assuming that shared tokens refer to similar meanings across languages. However, when word overlap is small, especially due to different writing systems, transfer is inhibited. In this paper, we define word-level information transfer pathways via word equivalence classes and rely on graph networks to fuse word embeddings across languages. Our experiments demonstrate the adv"},"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":"2305.14189","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T16:11:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"255bc3f51115e4dc6115370d61bbd1da998f460ac72f18f5c843a31daee57852","abstract_canon_sha256":"318a3ddf5fa46b67c4e163c7a9e6fe7d18c4fc956c90b7d85bca7c42fc6edcd3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:35:37.695197Z","signature_b64":"HjQjv8zz1vIRKoWQMfytahs3aSL+irJvfOolrOK8deW6mLiIfebLX566aRSFi+aomsSSVy6CAXNyZseqVyxnCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f16dd7f8f11f9045f435309f31a507048a0d222486639fc5a3e96e698635449","last_reissued_at":"2026-07-05T07:35:37.694778Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:35:37.694778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Shared Vocabulary: Increasing Representational Word Similarities across Languages for Multilingual Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Christof Monz, Di Wu","submitted_at":"2023-05-23T16:11:00Z","abstract_excerpt":"Using a vocabulary that is shared across languages is common practice in Multilingual Neural Machine Translation (MNMT). In addition to its simple design, shared tokens play an important role in positive knowledge transfer, assuming that shared tokens refer to similar meanings across languages. However, when word overlap is small, especially due to different writing systems, transfer is inhibited. In this paper, we define word-level information transfer pathways via word equivalence classes and rely on graph networks to fuse word embeddings across languages. Our experiments demonstrate the adv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14189","kind":"arxiv","version":3},"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/2305.14189/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":"2305.14189","created_at":"2026-07-05T07:35:37.694835+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14189v3","created_at":"2026-07-05T07:35:37.694835+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14189","created_at":"2026-07-05T07:35:37.694835+00:00"},{"alias_kind":"pith_short_12","alias_value":"P4LN274PCH4Q","created_at":"2026-07-05T07:35:37.694835+00:00"},{"alias_kind":"pith_short_16","alias_value":"P4LN274PCH4QIX2D","created_at":"2026-07-05T07:35:37.694835+00:00"},{"alias_kind":"pith_short_8","alias_value":"P4LN274P","created_at":"2026-07-05T07:35:37.694835+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.12183","citing_title":"Extend Adversarial Policy Against Neural Machine Translation via Unknown Token","ref_index":27,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB","json":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB.json","graph_json":"https://pith.science/api/pith-number/P4LN274PCH4QIX2DKME7GGSQOB/graph.json","events_json":"https://pith.science/api/pith-number/P4LN274PCH4QIX2DKME7GGSQOB/events.json","paper":"https://pith.science/paper/P4LN274P"},"agent_actions":{"view_html":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB","download_json":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB.json","view_paper":"https://pith.science/paper/P4LN274P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14189&json=true","fetch_graph":"https://pith.science/api/pith-number/P4LN274PCH4QIX2DKME7GGSQOB/graph.json","fetch_events":"https://pith.science/api/pith-number/P4LN274PCH4QIX2DKME7GGSQOB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB/action/storage_attestation","attest_author":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB/action/author_attestation","sign_citation":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB/action/citation_signature","submit_replication":"https://pith.science/pith/P4LN274PCH4QIX2DKME7GGSQOB/action/replication_record"}},"created_at":"2026-07-05T07:35:37.694835+00:00","updated_at":"2026-07-05T07:35:37.694835+00:00"}