{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XEVZEA2GT247BO2IFPZMKVMYAE","short_pith_number":"pith:XEVZEA2G","schema_version":"1.0","canonical_sha256":"b92b9203469eb9f0bb482bf2c555980126fd75d1819f1bbf025ed541bea201f7","source":{"kind":"arxiv","id":"2311.09205","version":1},"attestation_state":"computed","paper":{"title":"When Is Multilinguality a Curse? Language Modeling for 250 High- and Low-Resource Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin K. Bergen, Catherine Arnett, Tyler A. Chang, Zhuowen Tu","submitted_at":"2023-11-15T18:47:42Z","abstract_excerpt":"Multilingual language models are widely used to extend NLP systems to low-resource languages. However, concrete evidence for the effects of multilinguality on language modeling performance in individual languages remains scarce. Here, we pre-train over 10,000 monolingual and multilingual language models for over 250 languages, including multiple language families that are under-studied in NLP. We assess how language modeling performance in each language varies as a function of (1) monolingual dataset size, (2) added multilingual dataset size, (3) linguistic similarity of the added languages, a"},"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":"2311.09205","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-11-15T18:47:42Z","cross_cats_sorted":[],"title_canon_sha256":"b54d10f532a7382924d895c010be2a6c37c3d9b02ad9b22ede0a6dd5c63ddca8","abstract_canon_sha256":"dba8e761e9d6a274ee87990d059ba65c909c83920a76ec93d8587165f33a1da0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:13:10.206280Z","signature_b64":"bPwC9D4NpnjIfLLaJAnaN3TclCx9iQBc+pscPM3A36BJnazt7zEVlYiX3LV0JGmdL7A1E6FBF4LK17/Rj7kyBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b92b9203469eb9f0bb482bf2c555980126fd75d1819f1bbf025ed541bea201f7","last_reissued_at":"2026-07-05T07:13:10.205767Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:13:10.205767Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Is Multilinguality a Curse? Language Modeling for 250 High- and Low-Resource Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin K. Bergen, Catherine Arnett, Tyler A. Chang, Zhuowen Tu","submitted_at":"2023-11-15T18:47:42Z","abstract_excerpt":"Multilingual language models are widely used to extend NLP systems to low-resource languages. However, concrete evidence for the effects of multilinguality on language modeling performance in individual languages remains scarce. Here, we pre-train over 10,000 monolingual and multilingual language models for over 250 languages, including multiple language families that are under-studied in NLP. We assess how language modeling performance in each language varies as a function of (1) monolingual dataset size, (2) added multilingual dataset size, (3) linguistic similarity of the added languages, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.09205","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/2311.09205/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":"2311.09205","created_at":"2026-07-05T07:13:10.205824+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.09205v1","created_at":"2026-07-05T07:13:10.205824+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.09205","created_at":"2026-07-05T07:13:10.205824+00:00"},{"alias_kind":"pith_short_12","alias_value":"XEVZEA2GT247","created_at":"2026-07-05T07:13:10.205824+00:00"},{"alias_kind":"pith_short_16","alias_value":"XEVZEA2GT247BO2I","created_at":"2026-07-05T07:13:10.205824+00:00"},{"alias_kind":"pith_short_8","alias_value":"XEVZEA2G","created_at":"2026-07-05T07:13:10.205824+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.13225","citing_title":"Mix, Don't Tune: Bilingual Pre-Training Outperforms Hyperparameter Search in Data-Constrained Settings","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE","json":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE.json","graph_json":"https://pith.science/api/pith-number/XEVZEA2GT247BO2IFPZMKVMYAE/graph.json","events_json":"https://pith.science/api/pith-number/XEVZEA2GT247BO2IFPZMKVMYAE/events.json","paper":"https://pith.science/paper/XEVZEA2G"},"agent_actions":{"view_html":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE","download_json":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE.json","view_paper":"https://pith.science/paper/XEVZEA2G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.09205&json=true","fetch_graph":"https://pith.science/api/pith-number/XEVZEA2GT247BO2IFPZMKVMYAE/graph.json","fetch_events":"https://pith.science/api/pith-number/XEVZEA2GT247BO2IFPZMKVMYAE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE/action/storage_attestation","attest_author":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE/action/author_attestation","sign_citation":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE/action/citation_signature","submit_replication":"https://pith.science/pith/XEVZEA2GT247BO2IFPZMKVMYAE/action/replication_record"}},"created_at":"2026-07-05T07:13:10.205824+00:00","updated_at":"2026-07-05T07:13:10.205824+00:00"}