{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7CHGR3TLOGNOYX3VENHLQ3CC3B","short_pith_number":"pith:7CHGR3TL","schema_version":"1.0","canonical_sha256":"f88e68ee6b719aec5f75234eb86c42d877754b356a73f84df7b248e9e13e745d","source":{"kind":"arxiv","id":"2506.06751","version":2},"attestation_state":"computed","paper":{"title":"Geopolitical biases in LLMs: what are the \"good\" and the \"bad\" countries according to contemporary language models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexander Panchenko, Artyom Iudin, Dmitrii Korzh, Elena Tutubalina, Elvir Karimov, Ivan Lazichny, Ivan Oseledets, Mikhail Salnikov, Natalia Loukachevitch, Oleg Y. Rogov","submitted_at":"2025-06-07T10:45:17Z","abstract_excerpt":"This paper evaluates geopolitical biases in LLMs with respect to various countries though an analysis of their interpretation of historical events with conflicting national perspectives (USA, UK, USSR, and China). We introduce a novel dataset with neutral event descriptions and contrasting viewpoints from different countries. Our findings show significant geopolitical biases, with models favoring specific national narratives. Additionally, simple debiasing prompts had a limited effect in reducing these biases. Experiments with manipulated participant labels reveal models' sensitivity to attrib"},"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":"2506.06751","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-07T10:45:17Z","cross_cats_sorted":[],"title_canon_sha256":"2f429048eca4edf54b778ccdb6574b5d72a7c0e7d11dd86566381a347b0a05a7","abstract_canon_sha256":"2241364a4188494cf0f1262e2b4d9705a01bc0f0dd59eee24e610693cdf4e009"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:53.351554Z","signature_b64":"pkq5FA9HAwufZicB+8zTu6TGIRqda5fNMtaoTGj1i8fbruqquzCmSyB9prpTep7OC6hneFlsiTNYpLHvaoRYCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f88e68ee6b719aec5f75234eb86c42d877754b356a73f84df7b248e9e13e745d","last_reissued_at":"2026-07-05T11:24:53.351023Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:53.351023Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Geopolitical biases in LLMs: what are the \"good\" and the \"bad\" countries according to contemporary language models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexander Panchenko, Artyom Iudin, Dmitrii Korzh, Elena Tutubalina, Elvir Karimov, Ivan Lazichny, Ivan Oseledets, Mikhail Salnikov, Natalia Loukachevitch, Oleg Y. Rogov","submitted_at":"2025-06-07T10:45:17Z","abstract_excerpt":"This paper evaluates geopolitical biases in LLMs with respect to various countries though an analysis of their interpretation of historical events with conflicting national perspectives (USA, UK, USSR, and China). We introduce a novel dataset with neutral event descriptions and contrasting viewpoints from different countries. Our findings show significant geopolitical biases, with models favoring specific national narratives. Additionally, simple debiasing prompts had a limited effect in reducing these biases. Experiments with manipulated participant labels reveal models' sensitivity to attrib"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06751","kind":"arxiv","version":2},"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/2506.06751/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":"2506.06751","created_at":"2026-07-05T11:24:53.351086+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.06751v2","created_at":"2026-07-05T11:24:53.351086+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06751","created_at":"2026-07-05T11:24:53.351086+00:00"},{"alias_kind":"pith_short_12","alias_value":"7CHGR3TLOGNO","created_at":"2026-07-05T11:24:53.351086+00:00"},{"alias_kind":"pith_short_16","alias_value":"7CHGR3TLOGNOYX3V","created_at":"2026-07-05T11:24:53.351086+00:00"},{"alias_kind":"pith_short_8","alias_value":"7CHGR3TL","created_at":"2026-07-05T11:24:53.351086+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18124","citing_title":"Unintended Effects of Geographic Conditioning in Large Language Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22841","citing_title":"Strategic Coercion Within Alliances: The Greenland Sovereignty Game as an AI Stress Test","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22720","citing_title":"Can AI Make Conflicts Worse? An Alignment Failure in LLM Deployment Across Conflict Contexts","ref_index":4,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B","json":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B.json","graph_json":"https://pith.science/api/pith-number/7CHGR3TLOGNOYX3VENHLQ3CC3B/graph.json","events_json":"https://pith.science/api/pith-number/7CHGR3TLOGNOYX3VENHLQ3CC3B/events.json","paper":"https://pith.science/paper/7CHGR3TL"},"agent_actions":{"view_html":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B","download_json":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B.json","view_paper":"https://pith.science/paper/7CHGR3TL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.06751&json=true","fetch_graph":"https://pith.science/api/pith-number/7CHGR3TLOGNOYX3VENHLQ3CC3B/graph.json","fetch_events":"https://pith.science/api/pith-number/7CHGR3TLOGNOYX3VENHLQ3CC3B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B/action/storage_attestation","attest_author":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B/action/author_attestation","sign_citation":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B/action/citation_signature","submit_replication":"https://pith.science/pith/7CHGR3TLOGNOYX3VENHLQ3CC3B/action/replication_record"}},"created_at":"2026-07-05T11:24:53.351086+00:00","updated_at":"2026-07-05T11:24:53.351086+00:00"}