{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4BDIA2IK5IUKRFMEUEU2DZOGQ5","short_pith_number":"pith:4BDIA2IK","schema_version":"1.0","canonical_sha256":"e04680690aea28a89584a129a1e5c6874647ac7bf9986d02d482d160ad89ff1b","source":{"kind":"arxiv","id":"2507.16989","version":1},"attestation_state":"computed","paper":{"title":"Obscured but Not Erased: Evaluating Nationality Bias in LLMs via Name-Based Bias Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cristovao Iglesias, Devesh Batra, Giulio Pelosio, Greig Cowan, No\\'emie Bovey, Raad Khraishi, Robert Hankache","submitted_at":"2025-07-22T19:54:49Z","abstract_excerpt":"Large Language Models (LLMs) can exhibit latent biases towards specific nationalities even when explicit demographic markers are not present. In this work, we introduce a novel name-based benchmarking approach derived from the Bias Benchmark for QA (BBQ) dataset to investigate the impact of substituting explicit nationality labels with culturally indicative names, a scenario more reflective of real-world LLM applications. Our novel approach examines how this substitution affects both bias magnitude and accuracy across a spectrum of LLMs from industry leaders such as OpenAI, Google, and Anthrop"},"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":"2507.16989","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-22T19:54:49Z","cross_cats_sorted":[],"title_canon_sha256":"6e921ab0b6ad33606ded365380048b4a33d3f11331ffb9bba1569bc784d7adc7","abstract_canon_sha256":"0ff943886213da259167966ff56e27ee11141e074d627e3ea4b16562f8693849"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:41:59.830786Z","signature_b64":"FW71uZ/QZ3doRGmpmcPJmzgIjxqnjdG7kJSuax9JEP7gMmNm4CLGU0xrLqQxZCYZQm7ZPtvh6L+M+G625Y1zCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e04680690aea28a89584a129a1e5c6874647ac7bf9986d02d482d160ad89ff1b","last_reissued_at":"2026-07-05T11:41:59.830295Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:41:59.830295Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Obscured but Not Erased: Evaluating Nationality Bias in LLMs via Name-Based Bias Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cristovao Iglesias, Devesh Batra, Giulio Pelosio, Greig Cowan, No\\'emie Bovey, Raad Khraishi, Robert Hankache","submitted_at":"2025-07-22T19:54:49Z","abstract_excerpt":"Large Language Models (LLMs) can exhibit latent biases towards specific nationalities even when explicit demographic markers are not present. In this work, we introduce a novel name-based benchmarking approach derived from the Bias Benchmark for QA (BBQ) dataset to investigate the impact of substituting explicit nationality labels with culturally indicative names, a scenario more reflective of real-world LLM applications. Our novel approach examines how this substitution affects both bias magnitude and accuracy across a spectrum of LLMs from industry leaders such as OpenAI, Google, and Anthrop"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.16989","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/2507.16989/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":"2507.16989","created_at":"2026-07-05T11:41:59.830366+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.16989v1","created_at":"2026-07-05T11:41:59.830366+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.16989","created_at":"2026-07-05T11:41:59.830366+00:00"},{"alias_kind":"pith_short_12","alias_value":"4BDIA2IK5IUK","created_at":"2026-07-05T11:41:59.830366+00:00"},{"alias_kind":"pith_short_16","alias_value":"4BDIA2IK5IUKRFME","created_at":"2026-07-05T11:41:59.830366+00:00"},{"alias_kind":"pith_short_8","alias_value":"4BDIA2IK","created_at":"2026-07-05T11:41:59.830366+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/4BDIA2IK5IUKRFMEUEU2DZOGQ5","json":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5.json","graph_json":"https://pith.science/api/pith-number/4BDIA2IK5IUKRFMEUEU2DZOGQ5/graph.json","events_json":"https://pith.science/api/pith-number/4BDIA2IK5IUKRFMEUEU2DZOGQ5/events.json","paper":"https://pith.science/paper/4BDIA2IK"},"agent_actions":{"view_html":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5","download_json":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5.json","view_paper":"https://pith.science/paper/4BDIA2IK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.16989&json=true","fetch_graph":"https://pith.science/api/pith-number/4BDIA2IK5IUKRFMEUEU2DZOGQ5/graph.json","fetch_events":"https://pith.science/api/pith-number/4BDIA2IK5IUKRFMEUEU2DZOGQ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5/action/storage_attestation","attest_author":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5/action/author_attestation","sign_citation":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5/action/citation_signature","submit_replication":"https://pith.science/pith/4BDIA2IK5IUKRFMEUEU2DZOGQ5/action/replication_record"}},"created_at":"2026-07-05T11:41:59.830366+00:00","updated_at":"2026-07-05T11:41:59.830366+00:00"}