{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QN6YTJ74X2OGL7FSE35PYA56AY","short_pith_number":"pith:QN6YTJ74","schema_version":"1.0","canonical_sha256":"837d89a7fcbe9c65fcb226fafc03be061c16942961463ffdda0f455b719fe89f","source":{"kind":"arxiv","id":"2508.05525","version":1},"attestation_state":"computed","paper":{"title":"The World According to LLMs: How Geographic Origin Influences LLMs' Entity Deduction Capabilities","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ali Emami, Harsh Nishant Lalai, Jiaxin Pei, Raj Sanjay Shah, Sashank Varma, Yi-Chia Wang","submitted_at":"2025-08-07T15:53:30Z","abstract_excerpt":"Large Language Models (LLMs) have been extensively tuned to mitigate explicit biases, yet they often exhibit subtle implicit biases rooted in their pre-training data. Rather than directly probing LLMs with human-crafted questions that may trigger guardrails, we propose studying how models behave when they proactively ask questions themselves. The 20 Questions game, a multi-turn deduction task, serves as an ideal testbed for this purpose. We systematically evaluate geographic performance disparities in entity deduction using a new dataset, Geo20Q+, consisting of both notable people and cultural"},"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":"2508.05525","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-07T15:53:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2826f732408889125b8ffbfc4f796d3ebd2d5a44e92f9b4592d75449eca31667","abstract_canon_sha256":"4bd71648fb8c7306d89e53079bd063c9d9274b4f806081a50af28f75d787960a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:14.843557Z","signature_b64":"P+5CF2BtrnSCUJcsYEBCIIDjcT3Hvj07yjfkgwYcH7d+VUc/yHNDe7heG56rQvTUIl3o/jX2lmM1RXWRDaXYBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"837d89a7fcbe9c65fcb226fafc03be061c16942961463ffdda0f455b719fe89f","last_reissued_at":"2026-07-05T11:50:14.843084Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:14.843084Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The World According to LLMs: How Geographic Origin Influences LLMs' Entity Deduction Capabilities","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ali Emami, Harsh Nishant Lalai, Jiaxin Pei, Raj Sanjay Shah, Sashank Varma, Yi-Chia Wang","submitted_at":"2025-08-07T15:53:30Z","abstract_excerpt":"Large Language Models (LLMs) have been extensively tuned to mitigate explicit biases, yet they often exhibit subtle implicit biases rooted in their pre-training data. Rather than directly probing LLMs with human-crafted questions that may trigger guardrails, we propose studying how models behave when they proactively ask questions themselves. The 20 Questions game, a multi-turn deduction task, serves as an ideal testbed for this purpose. We systematically evaluate geographic performance disparities in entity deduction using a new dataset, Geo20Q+, consisting of both notable people and cultural"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05525","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/2508.05525/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":"2508.05525","created_at":"2026-07-05T11:50:14.843140+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.05525v1","created_at":"2026-07-05T11:50:14.843140+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05525","created_at":"2026-07-05T11:50:14.843140+00:00"},{"alias_kind":"pith_short_12","alias_value":"QN6YTJ74X2OG","created_at":"2026-07-05T11:50:14.843140+00:00"},{"alias_kind":"pith_short_16","alias_value":"QN6YTJ74X2OGL7FS","created_at":"2026-07-05T11:50:14.843140+00:00"},{"alias_kind":"pith_short_8","alias_value":"QN6YTJ74","created_at":"2026-07-05T11:50:14.843140+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.26099","citing_title":"Benchmarking Open-Weight Foundation Models for Global AI Technical Governance","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY","json":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY.json","graph_json":"https://pith.science/api/pith-number/QN6YTJ74X2OGL7FSE35PYA56AY/graph.json","events_json":"https://pith.science/api/pith-number/QN6YTJ74X2OGL7FSE35PYA56AY/events.json","paper":"https://pith.science/paper/QN6YTJ74"},"agent_actions":{"view_html":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY","download_json":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY.json","view_paper":"https://pith.science/paper/QN6YTJ74","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.05525&json=true","fetch_graph":"https://pith.science/api/pith-number/QN6YTJ74X2OGL7FSE35PYA56AY/graph.json","fetch_events":"https://pith.science/api/pith-number/QN6YTJ74X2OGL7FSE35PYA56AY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY/action/storage_attestation","attest_author":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY/action/author_attestation","sign_citation":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY/action/citation_signature","submit_replication":"https://pith.science/pith/QN6YTJ74X2OGL7FSE35PYA56AY/action/replication_record"}},"created_at":"2026-07-05T11:50:14.843140+00:00","updated_at":"2026-07-05T11:50:14.843140+00:00"}