{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:2FARITXWSPIRXGLMTTS4TRVXCF","short_pith_number":"pith:2FARITXW","schema_version":"1.0","canonical_sha256":"d141144ef693d11b996c9ce5c9c6b71140f367db218d055c725836ea35cb3b4f","source":{"kind":"arxiv","id":"2012.05958","version":1},"attestation_state":"computed","paper":{"title":"Multilingual Transfer Learning for QA Using Translation as Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Avirup Sil, Lin Pan, Mihaela Bornea, Radu Florian, Sara Rosenthal","submitted_at":"2020-12-10T20:29:34Z","abstract_excerpt":"Prior work on multilingual question answering has mostly focused on using large multilingual pre-trained language models (LM) to perform zero-shot language-wise learning: train a QA model on English and test on other languages. In this work, we explore strategies that improve cross-lingual transfer by bringing the multilingual embeddings closer in the semantic space. Our first strategy augments the original English training data with machine translation-generated data. This results in a corpus of multilingual silver-labeled QA pairs that is 14 times larger than the original training set. In ad"},"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":"2012.05958","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-12-10T20:29:34Z","cross_cats_sorted":[],"title_canon_sha256":"ca5f59eff708bc26c23b77ea0c635b2c9c6ebb333d01235f3263cd603c097076","abstract_canon_sha256":"01e73e0588849bff05e102c9a89d30cf9c8227cd4fed5a438d853f60b28d8261"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:58:45.363331Z","signature_b64":"7hmAwjOZgt6UVzPa4uzDXTHjVRoWIBQsn2V7HvzTz18G1hbYJIOWnhO+Svux15SRgg5gp+p4NCCjDix2dZbyBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d141144ef693d11b996c9ce5c9c6b71140f367db218d055c725836ea35cb3b4f","last_reissued_at":"2026-07-05T01:58:45.361853Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:58:45.361853Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilingual Transfer Learning for QA Using Translation as Data Augmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Avirup Sil, Lin Pan, Mihaela Bornea, Radu Florian, Sara Rosenthal","submitted_at":"2020-12-10T20:29:34Z","abstract_excerpt":"Prior work on multilingual question answering has mostly focused on using large multilingual pre-trained language models (LM) to perform zero-shot language-wise learning: train a QA model on English and test on other languages. In this work, we explore strategies that improve cross-lingual transfer by bringing the multilingual embeddings closer in the semantic space. Our first strategy augments the original English training data with machine translation-generated data. This results in a corpus of multilingual silver-labeled QA pairs that is 14 times larger than the original training set. In ad"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.05958","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/2012.05958/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":"2012.05958","created_at":"2026-07-05T01:58:45.362689+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.05958v1","created_at":"2026-07-05T01:58:45.362689+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.05958","created_at":"2026-07-05T01:58:45.362689+00:00"},{"alias_kind":"pith_short_12","alias_value":"2FARITXWSPIR","created_at":"2026-07-05T01:58:45.362689+00:00"},{"alias_kind":"pith_short_16","alias_value":"2FARITXWSPIRXGLM","created_at":"2026-07-05T01:58:45.362689+00:00"},{"alias_kind":"pith_short_8","alias_value":"2FARITXW","created_at":"2026-07-05T01:58:45.362689+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/2FARITXWSPIRXGLMTTS4TRVXCF","json":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF.json","graph_json":"https://pith.science/api/pith-number/2FARITXWSPIRXGLMTTS4TRVXCF/graph.json","events_json":"https://pith.science/api/pith-number/2FARITXWSPIRXGLMTTS4TRVXCF/events.json","paper":"https://pith.science/paper/2FARITXW"},"agent_actions":{"view_html":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF","download_json":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF.json","view_paper":"https://pith.science/paper/2FARITXW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.05958&json=true","fetch_graph":"https://pith.science/api/pith-number/2FARITXWSPIRXGLMTTS4TRVXCF/graph.json","fetch_events":"https://pith.science/api/pith-number/2FARITXWSPIRXGLMTTS4TRVXCF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF/action/storage_attestation","attest_author":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF/action/author_attestation","sign_citation":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF/action/citation_signature","submit_replication":"https://pith.science/pith/2FARITXWSPIRXGLMTTS4TRVXCF/action/replication_record"}},"created_at":"2026-07-05T01:58:45.362689+00:00","updated_at":"2026-07-05T01:58:45.362689+00:00"}