{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:5REURI35HW7TZ3MHVQVXCOB5AX","short_pith_number":"pith:5REURI35","schema_version":"1.0","canonical_sha256":"ec4948a37d3dbf3ced87ac2b71383d05f31cf2332494f81d99b5ff2c1d59e4de","source":{"kind":"arxiv","id":"2305.12182","version":2},"attestation_state":"computed","paper":{"title":"Glot500: Scaling Multilingual Corpora and Language Models to 500 Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amir Hossein Kargaran, Andr\\'e F. T. Martins, Ayyoob Imani, Chunlan Ma, Fran\\c{c}ois Yvon, Helmut Schmid, Hinrich Sch\\\"utze, Masoud Jalili Sabet, Nora Kassner, Peiqin Lin, Silvia Severini","submitted_at":"2023-05-20T12:26:41Z","abstract_excerpt":"The NLP community has mainly focused on scaling Large Language Models (LLMs) vertically, i.e., making them better for about 100 languages. We instead scale LLMs horizontally: we create, through continued pretraining, Glot500-m, an LLM that covers 511 predominantly low-resource languages. An important part of this effort is to collect and clean Glot500-c, a corpus that covers these 511 languages and allows us to train Glot500-m. We evaluate Glot500-m on five diverse tasks across these languages. We observe large improvements for both high-resource and low-resource languages compared to an XLM-R"},"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":"2305.12182","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-20T12:26:41Z","cross_cats_sorted":[],"title_canon_sha256":"2eadaf98b0a806d0965987715b0c306afd8ec671bcd9601492ac9bea9adc6c30","abstract_canon_sha256":"458030d49991b575a80f4af11c45b9470940bfd4d9c7d73f285035848e165f1a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:46:04.697707Z","signature_b64":"ddh2Rf7H2k4BngOxKF/MOXLnXCIecE/8zKbeqT9xCpHNftGzeTyhLxGwpz1IvSGQ0GvejimhPzle63uW43kuBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ec4948a37d3dbf3ced87ac2b71383d05f31cf2332494f81d99b5ff2c1d59e4de","last_reissued_at":"2026-07-05T06:46:04.697249Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:46:04.697249Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Glot500: Scaling Multilingual Corpora and Language Models to 500 Languages","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amir Hossein Kargaran, Andr\\'e F. T. Martins, Ayyoob Imani, Chunlan Ma, Fran\\c{c}ois Yvon, Helmut Schmid, Hinrich Sch\\\"utze, Masoud Jalili Sabet, Nora Kassner, Peiqin Lin, Silvia Severini","submitted_at":"2023-05-20T12:26:41Z","abstract_excerpt":"The NLP community has mainly focused on scaling Large Language Models (LLMs) vertically, i.e., making them better for about 100 languages. We instead scale LLMs horizontally: we create, through continued pretraining, Glot500-m, an LLM that covers 511 predominantly low-resource languages. An important part of this effort is to collect and clean Glot500-c, a corpus that covers these 511 languages and allows us to train Glot500-m. We evaluate Glot500-m on five diverse tasks across these languages. We observe large improvements for both high-resource and low-resource languages compared to an XLM-R"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.12182","kind":"arxiv","version":2},"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/2305.12182/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":"2305.12182","created_at":"2026-07-05T06:46:04.697308+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.12182v2","created_at":"2026-07-05T06:46:04.697308+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.12182","created_at":"2026-07-05T06:46:04.697308+00:00"},{"alias_kind":"pith_short_12","alias_value":"5REURI35HW7T","created_at":"2026-07-05T06:46:04.697308+00:00"},{"alias_kind":"pith_short_16","alias_value":"5REURI35HW7TZ3MH","created_at":"2026-07-05T06:46:04.697308+00:00"},{"alias_kind":"pith_short_8","alias_value":"5REURI35","created_at":"2026-07-05T06:46:04.697308+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/5REURI35HW7TZ3MHVQVXCOB5AX","json":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX.json","graph_json":"https://pith.science/api/pith-number/5REURI35HW7TZ3MHVQVXCOB5AX/graph.json","events_json":"https://pith.science/api/pith-number/5REURI35HW7TZ3MHVQVXCOB5AX/events.json","paper":"https://pith.science/paper/5REURI35"},"agent_actions":{"view_html":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX","download_json":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX.json","view_paper":"https://pith.science/paper/5REURI35","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.12182&json=true","fetch_graph":"https://pith.science/api/pith-number/5REURI35HW7TZ3MHVQVXCOB5AX/graph.json","fetch_events":"https://pith.science/api/pith-number/5REURI35HW7TZ3MHVQVXCOB5AX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX/action/storage_attestation","attest_author":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX/action/author_attestation","sign_citation":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX/action/citation_signature","submit_replication":"https://pith.science/pith/5REURI35HW7TZ3MHVQVXCOB5AX/action/replication_record"}},"created_at":"2026-07-05T06:46:04.697308+00:00","updated_at":"2026-07-05T06:46:04.697308+00:00"}