{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RLMSMLAIQ66T7CPWH7KG54ORXP","short_pith_number":"pith:RLMSMLAI","schema_version":"1.0","canonical_sha256":"8ad9262c0887bd3f89f63fd46ef1d1bbc309ee33d9a87f4db8458e53c48ccf35","source":{"kind":"arxiv","id":"2504.17052","version":4},"attestation_state":"computed","paper":{"title":"PReSS: An Automated Black-Box Framework for Evaluating Political Stance Stability in LLMs","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Kevin Esterling, Shariar Kabir, Yue Dong","submitted_at":"2025-04-23T19:00:39Z","abstract_excerpt":"Existing evaluations of political bias in large language models (LLMs) typically classify outputs as left- or right-leaning. We extend this perspective by examining how ideological tendencies vary across topics and how consistently models maintain their positions, a property we refer to as stability. To capture this dimension, we propose PReSS (Political Response Stability under Stress), an automated black-box framework that evaluates LLMs by jointly considering model and topic context, categorizing responses into four stance types: stable-left, unstable-left, stable-right, and unstable-right."},"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":"2504.17052","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-23T19:00:39Z","cross_cats_sorted":[],"title_canon_sha256":"fb294204bfdd06ddcad2812df8f36079681a844d3e00c3cccf81c17c9b0bca39","abstract_canon_sha256":"e0e1cb129bf090f69554fa42ea200663166ba8200a90f37306ddab689db4eea7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T00:21:31.005842Z","signature_b64":"CZW+yaWzp/8JnoQU62o6YZd4oznLOB81qWRGAjz+YVqhYsDDb+4NcxmMurpa7NUttP6AIIiyQXUqkV679/qLDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8ad9262c0887bd3f89f63fd46ef1d1bbc309ee33d9a87f4db8458e53c48ccf35","last_reissued_at":"2026-07-28T00:21:31.004918Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T00:21:31.004918Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PReSS: An Automated Black-Box Framework for Evaluating Political Stance Stability in LLMs","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Kevin Esterling, Shariar Kabir, Yue Dong","submitted_at":"2025-04-23T19:00:39Z","abstract_excerpt":"Existing evaluations of political bias in large language models (LLMs) typically classify outputs as left- or right-leaning. We extend this perspective by examining how ideological tendencies vary across topics and how consistently models maintain their positions, a property we refer to as stability. To capture this dimension, we propose PReSS (Political Response Stability under Stress), an automated black-box framework that evaluates LLMs by jointly considering model and topic context, categorizing responses into four stance types: stable-left, unstable-left, stable-right, and unstable-right."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.17052","kind":"arxiv","version":4},"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/2504.17052/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":"2504.17052","created_at":"2026-07-28T00:21:31.005343+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.17052v4","created_at":"2026-07-28T00:21:31.005343+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.17052","created_at":"2026-07-28T00:21:31.005343+00:00"},{"alias_kind":"pith_short_12","alias_value":"RLMSMLAIQ66T","created_at":"2026-07-28T00:21:31.005343+00:00"},{"alias_kind":"pith_short_16","alias_value":"RLMSMLAIQ66T7CPW","created_at":"2026-07-28T00:21:31.005343+00:00"},{"alias_kind":"pith_short_8","alias_value":"RLMSMLAI","created_at":"2026-07-28T00:21:31.005343+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.00048","citing_title":"Progressive in Principle, Centrist in Practice: LLM Political Bias Is Instrument-Dependent","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP","json":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP.json","graph_json":"https://pith.science/api/pith-number/RLMSMLAIQ66T7CPWH7KG54ORXP/graph.json","events_json":"https://pith.science/api/pith-number/RLMSMLAIQ66T7CPWH7KG54ORXP/events.json","paper":"https://pith.science/paper/RLMSMLAI"},"agent_actions":{"view_html":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP","download_json":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP.json","view_paper":"https://pith.science/paper/RLMSMLAI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.17052&json=true","fetch_graph":"https://pith.science/api/pith-number/RLMSMLAIQ66T7CPWH7KG54ORXP/graph.json","fetch_events":"https://pith.science/api/pith-number/RLMSMLAIQ66T7CPWH7KG54ORXP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP/action/storage_attestation","attest_author":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP/action/author_attestation","sign_citation":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP/action/citation_signature","submit_replication":"https://pith.science/pith/RLMSMLAIQ66T7CPWH7KG54ORXP/action/replication_record"}},"created_at":"2026-07-28T00:21:31.005343+00:00","updated_at":"2026-07-28T00:21:31.005343+00:00"}