{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:4CDT5YXMQVVF4FDIRH3JU5BN6J","short_pith_number":"pith:4CDT5YXM","schema_version":"1.0","canonical_sha256":"e0873ee2ec856a5e146889f69a742df270b4c6b115722ff2fef5f66c2d208441","source":{"kind":"arxiv","id":"2204.07580","version":2},"attestation_state":"computed","paper":{"title":"mGPT: Few-Shot Learners Go Multilingual","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alena Fenogenova, Anastasia Kozlova, Maria Tikhonova, Oleh Shliazhko, Tatiana Shavrina, Vladislav Mikhailov","submitted_at":"2022-04-15T13:02:33Z","abstract_excerpt":"Recent studies report that autoregressive language models can successfully solve many NLP tasks via zero- and few-shot learning paradigms, which opens up new possibilities for using the pre-trained language models. This paper introduces two autoregressive GPT-like models with 1.3 billion and 13 billion parameters trained on 60 languages from 25 language families using Wikipedia and Colossal Clean Crawled Corpus. We reproduce the GPT-3 architecture using GPT-2 sources and the sparse attention mechanism; Deepspeed and Megatron frameworks allow us to parallelize the training and inference steps e"},"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":"2204.07580","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-04-15T13:02:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8e163858d9d0b110bcc72dcb61a3e9f3b159704b9f692d637074532d0d6d6bcc","abstract_canon_sha256":"631b10ba1d5376eb60a70ced212974a55f90267266de21875bfbb5ca91c62a9e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:00:06.041015Z","signature_b64":"Db3W5nbpk7bqTT2gudds3IXen0hUR3wJS/3F8tvgIhBeGLSCVEs270BDz3NovGwAqVSkZ3xwl70B2dVUIgbkBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e0873ee2ec856a5e146889f69a742df270b4c6b115722ff2fef5f66c2d208441","last_reissued_at":"2026-07-05T07:00:06.040536Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:00:06.040536Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"mGPT: Few-Shot Learners Go Multilingual","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alena Fenogenova, Anastasia Kozlova, Maria Tikhonova, Oleh Shliazhko, Tatiana Shavrina, Vladislav Mikhailov","submitted_at":"2022-04-15T13:02:33Z","abstract_excerpt":"Recent studies report that autoregressive language models can successfully solve many NLP tasks via zero- and few-shot learning paradigms, which opens up new possibilities for using the pre-trained language models. This paper introduces two autoregressive GPT-like models with 1.3 billion and 13 billion parameters trained on 60 languages from 25 language families using Wikipedia and Colossal Clean Crawled Corpus. We reproduce the GPT-3 architecture using GPT-2 sources and the sparse attention mechanism; Deepspeed and Megatron frameworks allow us to parallelize the training and inference steps e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.07580","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/2204.07580/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":"2204.07580","created_at":"2026-07-05T07:00:06.040587+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.07580v2","created_at":"2026-07-05T07:00:06.040587+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.07580","created_at":"2026-07-05T07:00:06.040587+00:00"},{"alias_kind":"pith_short_12","alias_value":"4CDT5YXMQVVF","created_at":"2026-07-05T07:00:06.040587+00:00"},{"alias_kind":"pith_short_16","alias_value":"4CDT5YXMQVVF4FDI","created_at":"2026-07-05T07:00:06.040587+00:00"},{"alias_kind":"pith_short_8","alias_value":"4CDT5YXM","created_at":"2026-07-05T07:00:06.040587+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2211.05100","citing_title":"BLOOM: A 176B-Parameter Open-Access Multilingual Language Model","ref_index":183,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J","json":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J.json","graph_json":"https://pith.science/api/pith-number/4CDT5YXMQVVF4FDIRH3JU5BN6J/graph.json","events_json":"https://pith.science/api/pith-number/4CDT5YXMQVVF4FDIRH3JU5BN6J/events.json","paper":"https://pith.science/paper/4CDT5YXM"},"agent_actions":{"view_html":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J","download_json":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J.json","view_paper":"https://pith.science/paper/4CDT5YXM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.07580&json=true","fetch_graph":"https://pith.science/api/pith-number/4CDT5YXMQVVF4FDIRH3JU5BN6J/graph.json","fetch_events":"https://pith.science/api/pith-number/4CDT5YXMQVVF4FDIRH3JU5BN6J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J/action/storage_attestation","attest_author":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J/action/author_attestation","sign_citation":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J/action/citation_signature","submit_replication":"https://pith.science/pith/4CDT5YXMQVVF4FDIRH3JU5BN6J/action/replication_record"}},"created_at":"2026-07-05T07:00:06.040587+00:00","updated_at":"2026-07-05T07:00:06.040587+00:00"}