{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:LSCTTQXPB6GE7WBXPX4JKEM2AF","short_pith_number":"pith:LSCTTQXP","schema_version":"1.0","canonical_sha256":"5c8539c2ef0f8c4fd8377df895119a016ae32a7f920e0b8f1fad07bc63292c2d","source":{"kind":"arxiv","id":"2102.08357","version":1},"attestation_state":"computed","paper":{"title":"Revisiting Language Encoding in Learning Multilingual Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Di He, Guolin Ke, Kaiyuan Gao, Liwei Wang, Shengjie Luo, Shuxin Zheng, Tie-Yan Liu","submitted_at":"2021-02-16T18:47:10Z","abstract_excerpt":"Transformer has demonstrated its great power to learn contextual word representations for multiple languages in a single model. To process multilingual sentences in the model, a learnable vector is usually assigned to each language, which is called \"language embedding\". The language embedding can be either added to the word embedding or attached at the beginning of the sentence. It serves as a language-specific signal for the Transformer to capture contextual representations across languages. In this paper, we revisit the use of language embedding and identify several problems in the existing "},"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":"2102.08357","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-16T18:47:10Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a87f6ee6847d6cf831070535b405437c2704be02448a7377629d927ecee1e0ad","abstract_canon_sha256":"41ec7d95ab88f258adc35b17064a5f905f2d4a194f049e30301cf75c2cf36ec9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:15:48.934342Z","signature_b64":"cBvz4dlo8VRT0trLI0F7ZdUa6NIChtv4DZZxC9KEGKCFWxVuEEo1psq4JNq/88ZEIH4Cr3/Ce4mmD+vkcd+eBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5c8539c2ef0f8c4fd8377df895119a016ae32a7f920e0b8f1fad07bc63292c2d","last_reissued_at":"2026-07-05T02:15:48.933868Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:15:48.933868Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Revisiting Language Encoding in Learning Multilingual Representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Di He, Guolin Ke, Kaiyuan Gao, Liwei Wang, Shengjie Luo, Shuxin Zheng, Tie-Yan Liu","submitted_at":"2021-02-16T18:47:10Z","abstract_excerpt":"Transformer has demonstrated its great power to learn contextual word representations for multiple languages in a single model. To process multilingual sentences in the model, a learnable vector is usually assigned to each language, which is called \"language embedding\". The language embedding can be either added to the word embedding or attached at the beginning of the sentence. It serves as a language-specific signal for the Transformer to capture contextual representations across languages. In this paper, we revisit the use of language embedding and identify several problems in the existing "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.08357","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/2102.08357/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":"2102.08357","created_at":"2026-07-05T02:15:48.933925+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.08357v1","created_at":"2026-07-05T02:15:48.933925+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.08357","created_at":"2026-07-05T02:15:48.933925+00:00"},{"alias_kind":"pith_short_12","alias_value":"LSCTTQXPB6GE","created_at":"2026-07-05T02:15:48.933925+00:00"},{"alias_kind":"pith_short_16","alias_value":"LSCTTQXPB6GE7WBX","created_at":"2026-07-05T02:15:48.933925+00:00"},{"alias_kind":"pith_short_8","alias_value":"LSCTTQXP","created_at":"2026-07-05T02:15:48.933925+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/LSCTTQXPB6GE7WBXPX4JKEM2AF","json":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF.json","graph_json":"https://pith.science/api/pith-number/LSCTTQXPB6GE7WBXPX4JKEM2AF/graph.json","events_json":"https://pith.science/api/pith-number/LSCTTQXPB6GE7WBXPX4JKEM2AF/events.json","paper":"https://pith.science/paper/LSCTTQXP"},"agent_actions":{"view_html":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF","download_json":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF.json","view_paper":"https://pith.science/paper/LSCTTQXP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.08357&json=true","fetch_graph":"https://pith.science/api/pith-number/LSCTTQXPB6GE7WBXPX4JKEM2AF/graph.json","fetch_events":"https://pith.science/api/pith-number/LSCTTQXPB6GE7WBXPX4JKEM2AF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF/action/storage_attestation","attest_author":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF/action/author_attestation","sign_citation":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF/action/citation_signature","submit_replication":"https://pith.science/pith/LSCTTQXPB6GE7WBXPX4JKEM2AF/action/replication_record"}},"created_at":"2026-07-05T02:15:48.933925+00:00","updated_at":"2026-07-05T02:15:48.933925+00:00"}