{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:AQQWSTSHXBV7XHARIHJUZZLUCB","short_pith_number":"pith:AQQWSTSH","schema_version":"1.0","canonical_sha256":"0421694e47b86bfb9c1141d34ce5741074bfdf88f863ae8dde4b4a8b760c7023","source":{"kind":"arxiv","id":"2105.00572","version":1},"attestation_state":"computed","paper":{"title":"Larger-Scale Transformers for Multilingual Masked Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexis Conneau, Giri Anantharaman, Jingfei Du, Myle Ott, Naman Goyal","submitted_at":"2021-05-02T23:15:02Z","abstract_excerpt":"Recent work has demonstrated the effectiveness of cross-lingual language model pretraining for cross-lingual understanding. In this study, we present the results of two larger multilingual masked language models, with 3.5B and 10.7B parameters. Our two new models dubbed XLM-R XL and XLM-R XXL outperform XLM-R by 1.8% and 2.4% average accuracy on XNLI. Our model also outperforms the RoBERTa-Large model on several English tasks of the GLUE benchmark by 0.3% on average while handling 99 more languages. This suggests pretrained models with larger capacity may obtain both strong performance on high"},"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":"2105.00572","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-05-02T23:15:02Z","cross_cats_sorted":[],"title_canon_sha256":"654d54a90add4e13e26e4d9755992467edb7a0d30eaee0bd7eabcf916765ae5b","abstract_canon_sha256":"81bd879192e20a739347997f113647877dfe395f696e91cb648c113c74ec5223"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:36:55.590000Z","signature_b64":"R9k8zAXQYL9a/9qrRWY/PLs5AmHYUt2Mg9i+rMztm6fOhe0W+sBCiGKauZPmj9JASDv7J4MNwDN68R9IwReOCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0421694e47b86bfb9c1141d34ce5741074bfdf88f863ae8dde4b4a8b760c7023","last_reissued_at":"2026-07-05T02:36:55.589521Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:36:55.589521Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Larger-Scale Transformers for Multilingual Masked Language Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexis Conneau, Giri Anantharaman, Jingfei Du, Myle Ott, Naman Goyal","submitted_at":"2021-05-02T23:15:02Z","abstract_excerpt":"Recent work has demonstrated the effectiveness of cross-lingual language model pretraining for cross-lingual understanding. In this study, we present the results of two larger multilingual masked language models, with 3.5B and 10.7B parameters. Our two new models dubbed XLM-R XL and XLM-R XXL outperform XLM-R by 1.8% and 2.4% average accuracy on XNLI. Our model also outperforms the RoBERTa-Large model on several English tasks of the GLUE benchmark by 0.3% on average while handling 99 more languages. This suggests pretrained models with larger capacity may obtain both strong performance on high"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.00572","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/2105.00572/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":"2105.00572","created_at":"2026-07-05T02:36:55.589578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.00572v1","created_at":"2026-07-05T02:36:55.589578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.00572","created_at":"2026-07-05T02:36:55.589578+00:00"},{"alias_kind":"pith_short_12","alias_value":"AQQWSTSHXBV7","created_at":"2026-07-05T02:36:55.589578+00:00"},{"alias_kind":"pith_short_16","alias_value":"AQQWSTSHXBV7XHAR","created_at":"2026-07-05T02:36:55.589578+00:00"},{"alias_kind":"pith_short_8","alias_value":"AQQWSTSH","created_at":"2026-07-05T02:36:55.589578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26613","citing_title":"EcoTable: Cost-effective Table Integration in Data Lakes for Natural Language Queries","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2205.01068","citing_title":"OPT: Open Pre-trained Transformer Language Models","ref_index":169,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB","json":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB.json","graph_json":"https://pith.science/api/pith-number/AQQWSTSHXBV7XHARIHJUZZLUCB/graph.json","events_json":"https://pith.science/api/pith-number/AQQWSTSHXBV7XHARIHJUZZLUCB/events.json","paper":"https://pith.science/paper/AQQWSTSH"},"agent_actions":{"view_html":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB","download_json":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB.json","view_paper":"https://pith.science/paper/AQQWSTSH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.00572&json=true","fetch_graph":"https://pith.science/api/pith-number/AQQWSTSHXBV7XHARIHJUZZLUCB/graph.json","fetch_events":"https://pith.science/api/pith-number/AQQWSTSHXBV7XHARIHJUZZLUCB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB/action/storage_attestation","attest_author":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB/action/author_attestation","sign_citation":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB/action/citation_signature","submit_replication":"https://pith.science/pith/AQQWSTSHXBV7XHARIHJUZZLUCB/action/replication_record"}},"created_at":"2026-07-05T02:36:55.589578+00:00","updated_at":"2026-07-05T02:36:55.589578+00:00"}