{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:J4SLE43ODCP5LF7UVJ5LNNZOKX","short_pith_number":"pith:J4SLE43O","schema_version":"1.0","canonical_sha256":"4f24b2736e189fd597f4aa7ab6b72e55d80ddb2d05dd5d78d21d1da1e37552dc","source":{"kind":"arxiv","id":"2406.12822","version":3},"attestation_state":"computed","paper":{"title":"Is It Good Data for Multilingual Instruction Tuning or Just Bad Multilingual Evaluation for Large Language Models?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Barry Haddow, Pinzhen Chen, Simon Yu, Zhicheng Guo","submitted_at":"2024-06-18T17:43:47Z","abstract_excerpt":"Multilingual large language models are designed, claimed, and expected to cater to speakers of varied languages. We hypothesise that the current practices of fine-tuning and evaluating these models may not perfectly align with this objective owing to a heavy reliance on translation, which cannot cover language-specific knowledge but can introduce translation defects. It remains unknown whether the nature of the instruction data has an impact on the model output; conversely, it is questionable whether translated test sets can capture such nuances. Due to the often coupled practices of using tra"},"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":"2406.12822","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-18T17:43:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f7a82d4b0eaced5a104d73001c1a7ae32ac1ac6e05cbc791b0608083ed209b04","abstract_canon_sha256":"c53899d5f39152bbc289b4782602021cffcca7032cc4b8791d22b4a3324693f4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:12:07.494464Z","signature_b64":"AQxzLkNxpWRxAEVm8jGX+kN12Gy462d3UPgKVijeCdlJnwqhCi5v0MKbQwj+TXRY3BEMkx1zSadxPVkyJTOMDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f24b2736e189fd597f4aa7ab6b72e55d80ddb2d05dd5d78d21d1da1e37552dc","last_reissued_at":"2026-07-05T09:12:07.493978Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:12:07.493978Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is It Good Data for Multilingual Instruction Tuning or Just Bad Multilingual Evaluation for Large Language Models?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Barry Haddow, Pinzhen Chen, Simon Yu, Zhicheng Guo","submitted_at":"2024-06-18T17:43:47Z","abstract_excerpt":"Multilingual large language models are designed, claimed, and expected to cater to speakers of varied languages. We hypothesise that the current practices of fine-tuning and evaluating these models may not perfectly align with this objective owing to a heavy reliance on translation, which cannot cover language-specific knowledge but can introduce translation defects. It remains unknown whether the nature of the instruction data has an impact on the model output; conversely, it is questionable whether translated test sets can capture such nuances. Due to the often coupled practices of using tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.12822","kind":"arxiv","version":3},"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/2406.12822/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":"2406.12822","created_at":"2026-07-05T09:12:07.494036+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.12822v3","created_at":"2026-07-05T09:12:07.494036+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.12822","created_at":"2026-07-05T09:12:07.494036+00:00"},{"alias_kind":"pith_short_12","alias_value":"J4SLE43ODCP5","created_at":"2026-07-05T09:12:07.494036+00:00"},{"alias_kind":"pith_short_16","alias_value":"J4SLE43ODCP5LF7U","created_at":"2026-07-05T09:12:07.494036+00:00"},{"alias_kind":"pith_short_8","alias_value":"J4SLE43O","created_at":"2026-07-05T09:12:07.494036+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.12674","citing_title":"Train More Parameters But Mind Their Placement: Insights into Language Adaptation with PEFT","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX","json":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX.json","graph_json":"https://pith.science/api/pith-number/J4SLE43ODCP5LF7UVJ5LNNZOKX/graph.json","events_json":"https://pith.science/api/pith-number/J4SLE43ODCP5LF7UVJ5LNNZOKX/events.json","paper":"https://pith.science/paper/J4SLE43O"},"agent_actions":{"view_html":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX","download_json":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX.json","view_paper":"https://pith.science/paper/J4SLE43O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.12822&json=true","fetch_graph":"https://pith.science/api/pith-number/J4SLE43ODCP5LF7UVJ5LNNZOKX/graph.json","fetch_events":"https://pith.science/api/pith-number/J4SLE43ODCP5LF7UVJ5LNNZOKX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX/action/storage_attestation","attest_author":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX/action/author_attestation","sign_citation":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX/action/citation_signature","submit_replication":"https://pith.science/pith/J4SLE43ODCP5LF7UVJ5LNNZOKX/action/replication_record"}},"created_at":"2026-07-05T09:12:07.494036+00:00","updated_at":"2026-07-05T09:12:07.494036+00:00"}