{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VPGFCNQRCIPHPMWSFY5EMVYPVB","short_pith_number":"pith:VPGFCNQR","schema_version":"1.0","canonical_sha256":"abcc513611121e77b2d22e3a46570fa854be93a59b86958fbc4ff2de02ae57f6","source":{"kind":"arxiv","id":"2501.08322","version":1},"attestation_state":"computed","paper":{"title":"Exploring Robustness of Multilingual LLMs on Real-World Noisy Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Akbar Karimi, Amirhossein Aliakbarzadeh, Lucie Flek","submitted_at":"2025-01-14T18:55:35Z","abstract_excerpt":"Large Language Models (LLMs) are trained on Web data that might contain spelling errors made by humans. But do they become robust to similar real-world noise? In this paper, we investigate the effect of real-world spelling mistakes on the performance of 9 language models, with parameters ranging from 0.2B to 13B, in 3 different NLP tasks, namely Natural Language Inference (NLI), Name Entity Recognition (NER), and Intent Classification (IC). We perform our experiments on 6 different languages and build a dictionary of real-world noise for them using the Wikipedia edit history. We show that the "},"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":"2501.08322","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-14T18:55:35Z","cross_cats_sorted":[],"title_canon_sha256":"3c466309a158d5d0fbbb9363dda0a3ad303d1c3db8cca3fbee10393449a2e0e8","abstract_canon_sha256":"6861e4153d428f368bcd0136fe4ab6bb9d3766bfc9341d76d4734dc627383a82"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:00:59.488692Z","signature_b64":"/N/dLfofLf5DUZuAnzTmKjj3ywrga48bb0otrQjPV/4eQ73KRmAIOfbcScxGBYwjYCQC4n8C/SiA7ELQB2yUCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abcc513611121e77b2d22e3a46570fa854be93a59b86958fbc4ff2de02ae57f6","last_reissued_at":"2026-07-05T10:00:59.488256Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:00:59.488256Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exploring Robustness of Multilingual LLMs on Real-World Noisy Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Akbar Karimi, Amirhossein Aliakbarzadeh, Lucie Flek","submitted_at":"2025-01-14T18:55:35Z","abstract_excerpt":"Large Language Models (LLMs) are trained on Web data that might contain spelling errors made by humans. But do they become robust to similar real-world noise? In this paper, we investigate the effect of real-world spelling mistakes on the performance of 9 language models, with parameters ranging from 0.2B to 13B, in 3 different NLP tasks, namely Natural Language Inference (NLI), Name Entity Recognition (NER), and Intent Classification (IC). We perform our experiments on 6 different languages and build a dictionary of real-world noise for them using the Wikipedia edit history. We show that the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.08322","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/2501.08322/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":"2501.08322","created_at":"2026-07-05T10:00:59.488320+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.08322v1","created_at":"2026-07-05T10:00:59.488320+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.08322","created_at":"2026-07-05T10:00:59.488320+00:00"},{"alias_kind":"pith_short_12","alias_value":"VPGFCNQRCIPH","created_at":"2026-07-05T10:00:59.488320+00:00"},{"alias_kind":"pith_short_16","alias_value":"VPGFCNQRCIPHPMWS","created_at":"2026-07-05T10:00:59.488320+00:00"},{"alias_kind":"pith_short_8","alias_value":"VPGFCNQR","created_at":"2026-07-05T10:00:59.488320+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/VPGFCNQRCIPHPMWSFY5EMVYPVB","json":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB.json","graph_json":"https://pith.science/api/pith-number/VPGFCNQRCIPHPMWSFY5EMVYPVB/graph.json","events_json":"https://pith.science/api/pith-number/VPGFCNQRCIPHPMWSFY5EMVYPVB/events.json","paper":"https://pith.science/paper/VPGFCNQR"},"agent_actions":{"view_html":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB","download_json":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB.json","view_paper":"https://pith.science/paper/VPGFCNQR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.08322&json=true","fetch_graph":"https://pith.science/api/pith-number/VPGFCNQRCIPHPMWSFY5EMVYPVB/graph.json","fetch_events":"https://pith.science/api/pith-number/VPGFCNQRCIPHPMWSFY5EMVYPVB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB/action/storage_attestation","attest_author":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB/action/author_attestation","sign_citation":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB/action/citation_signature","submit_replication":"https://pith.science/pith/VPGFCNQRCIPHPMWSFY5EMVYPVB/action/replication_record"}},"created_at":"2026-07-05T10:00:59.488320+00:00","updated_at":"2026-07-05T10:00:59.488320+00:00"}