{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KCXMEWVBNZHIW5S4FNMRDYV72T","short_pith_number":"pith:KCXMEWVB","schema_version":"1.0","canonical_sha256":"50aec25aa16e4e8b765c2b5911e2bfd4f4900facd4da0461fd0bbb895b652575","source":{"kind":"arxiv","id":"2504.17720","version":2},"attestation_state":"computed","paper":{"title":"Multilingual Performance Biases of Large Language Models in Education","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Donya Rooein, Mrinmaya Sachan, Sankalan Pal Chowdhury, Vansh Gupta, Vil\\'em Zouhar","submitted_at":"2025-04-24T16:32:31Z","abstract_excerpt":"Large language models (LLMs) are increasingly being adopted in educational settings. These applications expand beyond English, though current LLMs remain primarily English-centric. In this work, we ascertain if their use in education settings in non-English languages is warranted. We evaluated the performance of popular LLMs on four educational tasks: identifying student misconceptions, providing targeted feedback, interactive tutoring, and grading translations in eight languages (Mandarin, Hindi, Arabic, German, Farsi, Telugu, Ukrainian, Czech) in addition to English. We find that the perform"},"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":"2504.17720","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-24T16:32:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7b7184dc80cb586e25ee25cc313d0e94fdb3ca9c12086a4adc8b0bfac86e21a5","abstract_canon_sha256":"73cf7c8c2eeee15f1f83d7e20332efbdb98c1d4858249f9179dea71149ddf9b8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:48:48.877078Z","signature_b64":"myFeFqwcEJrw+kDyjVXOKlhH6a62j41AUVj9fEyPBaccDcZmWOlyTqBcyZzvKG0NFUoRsUlTcjGdTDfh619VDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50aec25aa16e4e8b765c2b5911e2bfd4f4900facd4da0461fd0bbb895b652575","last_reissued_at":"2026-07-05T11:48:48.876577Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:48:48.876577Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilingual Performance Biases of Large Language Models in Education","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Donya Rooein, Mrinmaya Sachan, Sankalan Pal Chowdhury, Vansh Gupta, Vil\\'em Zouhar","submitted_at":"2025-04-24T16:32:31Z","abstract_excerpt":"Large language models (LLMs) are increasingly being adopted in educational settings. These applications expand beyond English, though current LLMs remain primarily English-centric. In this work, we ascertain if their use in education settings in non-English languages is warranted. We evaluated the performance of popular LLMs on four educational tasks: identifying student misconceptions, providing targeted feedback, interactive tutoring, and grading translations in eight languages (Mandarin, Hindi, Arabic, German, Farsi, Telugu, Ukrainian, Czech) in addition to English. We find that the perform"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.17720","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/2504.17720/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":"2504.17720","created_at":"2026-07-05T11:48:48.876638+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.17720v2","created_at":"2026-07-05T11:48:48.876638+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.17720","created_at":"2026-07-05T11:48:48.876638+00:00"},{"alias_kind":"pith_short_12","alias_value":"KCXMEWVBNZHI","created_at":"2026-07-05T11:48:48.876638+00:00"},{"alias_kind":"pith_short_16","alias_value":"KCXMEWVBNZHIW5S4","created_at":"2026-07-05T11:48:48.876638+00:00"},{"alias_kind":"pith_short_8","alias_value":"KCXMEWVB","created_at":"2026-07-05T11:48:48.876638+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.15835","citing_title":"Alvorada-Bench: Can Language Models Solve Brazilian University Entrance Exams?","ref_index":3,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T","json":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T.json","graph_json":"https://pith.science/api/pith-number/KCXMEWVBNZHIW5S4FNMRDYV72T/graph.json","events_json":"https://pith.science/api/pith-number/KCXMEWVBNZHIW5S4FNMRDYV72T/events.json","paper":"https://pith.science/paper/KCXMEWVB"},"agent_actions":{"view_html":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T","download_json":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T.json","view_paper":"https://pith.science/paper/KCXMEWVB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.17720&json=true","fetch_graph":"https://pith.science/api/pith-number/KCXMEWVBNZHIW5S4FNMRDYV72T/graph.json","fetch_events":"https://pith.science/api/pith-number/KCXMEWVBNZHIW5S4FNMRDYV72T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T/action/storage_attestation","attest_author":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T/action/author_attestation","sign_citation":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T/action/citation_signature","submit_replication":"https://pith.science/pith/KCXMEWVBNZHIW5S4FNMRDYV72T/action/replication_record"}},"created_at":"2026-07-05T11:48:48.876638+00:00","updated_at":"2026-07-05T11:48:48.876638+00:00"}