{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2KT5OONIELNQLNYJKUW3DOFBSN","short_pith_number":"pith:2KT5OONI","schema_version":"1.0","canonical_sha256":"d2a7d739a822db05b709552db1b8a19361d0375c64ab54eff59ffe21ff54c7c0","source":{"kind":"arxiv","id":"2501.04662","version":1},"attestation_state":"computed","paper":{"title":"On The Origin of Cultural Biases in Language Models: From Pre-training Data to Linguistic Phenomena","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Tarek Naous, Wei Xu","submitted_at":"2025-01-08T18:15:47Z","abstract_excerpt":"Language Models (LMs) have been shown to exhibit a strong preference towards entities associated with Western culture when operating in non-Western languages. In this paper, we aim to uncover the origins of entity-related cultural biases in LMs by analyzing several contributing factors, including the representation of entities in pre-training data and the impact of variations in linguistic phenomena across languages. We introduce CAMeL-2, a parallel Arabic-English benchmark of 58,086 entities associated with Arab and Western cultures and 367 masked natural contexts for entities. Our evaluation"},"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":"2501.04662","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-08T18:15:47Z","cross_cats_sorted":[],"title_canon_sha256":"f823ff74e7976baf3e0fad51f2fd5dcc9604b614a2cc7ae6fde10ee810bc43c1","abstract_canon_sha256":"7b232907de95907985c0fb7c1b93ea5272a9a1d421460e37af350e74c9c8c110"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:58:38.581812Z","signature_b64":"HXJ17OORVWAatjZd8touDOtD5PafA344zCdt71H/K+y9SqLJQG4gJULNtG9pmo+/kUa0oUsZHP+YD3fF9AmpCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2a7d739a822db05b709552db1b8a19361d0375c64ab54eff59ffe21ff54c7c0","last_reissued_at":"2026-07-05T09:58:38.581327Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:58:38.581327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On The Origin of Cultural Biases in Language Models: From Pre-training Data to Linguistic Phenomena","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Tarek Naous, Wei Xu","submitted_at":"2025-01-08T18:15:47Z","abstract_excerpt":"Language Models (LMs) have been shown to exhibit a strong preference towards entities associated with Western culture when operating in non-Western languages. In this paper, we aim to uncover the origins of entity-related cultural biases in LMs by analyzing several contributing factors, including the representation of entities in pre-training data and the impact of variations in linguistic phenomena across languages. We introduce CAMeL-2, a parallel Arabic-English benchmark of 58,086 entities associated with Arab and Western cultures and 367 masked natural contexts for entities. Our evaluation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.04662","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/2501.04662/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":"2501.04662","created_at":"2026-07-05T09:58:38.581387+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.04662v1","created_at":"2026-07-05T09:58:38.581387+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.04662","created_at":"2026-07-05T09:58:38.581387+00:00"},{"alias_kind":"pith_short_12","alias_value":"2KT5OONIELNQ","created_at":"2026-07-05T09:58:38.581387+00:00"},{"alias_kind":"pith_short_16","alias_value":"2KT5OONIELNQLNYJ","created_at":"2026-07-05T09:58:38.581387+00:00"},{"alias_kind":"pith_short_8","alias_value":"2KT5OONI","created_at":"2026-07-05T09:58:38.581387+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.21476","citing_title":"Igniting Creative Writing in Small Language Models: LLM-as-a-Judge versus Multi-Agent Refined Rewards","ref_index":2024,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN","json":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN.json","graph_json":"https://pith.science/api/pith-number/2KT5OONIELNQLNYJKUW3DOFBSN/graph.json","events_json":"https://pith.science/api/pith-number/2KT5OONIELNQLNYJKUW3DOFBSN/events.json","paper":"https://pith.science/paper/2KT5OONI"},"agent_actions":{"view_html":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN","download_json":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN.json","view_paper":"https://pith.science/paper/2KT5OONI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.04662&json=true","fetch_graph":"https://pith.science/api/pith-number/2KT5OONIELNQLNYJKUW3DOFBSN/graph.json","fetch_events":"https://pith.science/api/pith-number/2KT5OONIELNQLNYJKUW3DOFBSN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN/action/storage_attestation","attest_author":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN/action/author_attestation","sign_citation":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN/action/citation_signature","submit_replication":"https://pith.science/pith/2KT5OONIELNQLNYJKUW3DOFBSN/action/replication_record"}},"created_at":"2026-07-05T09:58:38.581387+00:00","updated_at":"2026-07-05T09:58:38.581387+00:00"}