{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:OX4YZRUKD43NBJWMHZOHKADRC3","short_pith_number":"pith:OX4YZRUK","schema_version":"1.0","canonical_sha256":"75f98cc68a1f36d0a6cc3e5c75007116ea10ed1445232a9eac9163d71e49f977","source":{"kind":"arxiv","id":"2405.06058","version":2},"attestation_state":"computed","paper":{"title":"Large Language Models Show Human-like Social Desirability Biases in Survey Responses","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CY","cs.HC"],"primary_cat":"cs.AI","authors_text":"Aadesh Salecha, Jo\\~ao Sedoc, Johannes C. Eichstaedt, Lyle H. Ungar, Molly E. Ireland, Shashanka Subrahmanya","submitted_at":"2024-05-09T19:02:53Z","abstract_excerpt":"As Large Language Models (LLMs) become widely used to model and simulate human behavior, understanding their biases becomes critical. We developed an experimental framework using Big Five personality surveys and uncovered a previously undetected social desirability bias in a wide range of LLMs. By systematically varying the number of questions LLMs were exposed to, we demonstrate their ability to infer when they are being evaluated. When personality evaluation is inferred, LLMs skew their scores towards the desirable ends of trait dimensions (i.e., increased extraversion, decreased neuroticism"},"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":"2405.06058","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-05-09T19:02:53Z","cross_cats_sorted":["cs.CL","cs.CY","cs.HC"],"title_canon_sha256":"780a1a9dc31596927e2d97744b2128115412122e49f20574b2d004a96bbdefd0","abstract_canon_sha256":"2874579d6ffe71d17a014d922a468083b57d6277abc53a16d85bea70bc367628"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:38:49.134305Z","signature_b64":"AK1ATtL/x3ifIrakWxUMranAXfhzOY70ev3j1qLZ4o5hOwBXEXCGOdt7NCj2aBvahnNISyWwP71Bt2SP+84VBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"75f98cc68a1f36d0a6cc3e5c75007116ea10ed1445232a9eac9163d71e49f977","last_reissued_at":"2026-07-05T09:38:49.133820Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:38:49.133820Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models Show Human-like Social Desirability Biases in Survey Responses","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CY","cs.HC"],"primary_cat":"cs.AI","authors_text":"Aadesh Salecha, Jo\\~ao Sedoc, Johannes C. Eichstaedt, Lyle H. Ungar, Molly E. Ireland, Shashanka Subrahmanya","submitted_at":"2024-05-09T19:02:53Z","abstract_excerpt":"As Large Language Models (LLMs) become widely used to model and simulate human behavior, understanding their biases becomes critical. We developed an experimental framework using Big Five personality surveys and uncovered a previously undetected social desirability bias in a wide range of LLMs. By systematically varying the number of questions LLMs were exposed to, we demonstrate their ability to infer when they are being evaluated. When personality evaluation is inferred, LLMs skew their scores towards the desirable ends of trait dimensions (i.e., increased extraversion, decreased neuroticism"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.06058","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/2405.06058/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":"2405.06058","created_at":"2026-07-05T09:38:49.133877+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.06058v2","created_at":"2026-07-05T09:38:49.133877+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.06058","created_at":"2026-07-05T09:38:49.133877+00:00"},{"alias_kind":"pith_short_12","alias_value":"OX4YZRUKD43N","created_at":"2026-07-05T09:38:49.133877+00:00"},{"alias_kind":"pith_short_16","alias_value":"OX4YZRUKD43NBJWM","created_at":"2026-07-05T09:38:49.133877+00:00"},{"alias_kind":"pith_short_8","alias_value":"OX4YZRUK","created_at":"2026-07-05T09:38:49.133877+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11875","citing_title":"I Understand How You Feel: Enhancing Deeper Emotional Support Through Multilingual Emotional Validation in Dialogue System","ref_index":95,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16996","citing_title":"Evaluation Drift in LLM Personality Induction: Are We Moving the Goalpost?","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06196","citing_title":"The Granularity Axis: A Micro-to-Macro Latent Direction for Social Roles in Language Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05080","citing_title":"The Pinocchio Dimension: Phenomenality of Experience as the Primary Axis of LLM Psychometric Differences","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3","json":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3.json","graph_json":"https://pith.science/api/pith-number/OX4YZRUKD43NBJWMHZOHKADRC3/graph.json","events_json":"https://pith.science/api/pith-number/OX4YZRUKD43NBJWMHZOHKADRC3/events.json","paper":"https://pith.science/paper/OX4YZRUK"},"agent_actions":{"view_html":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3","download_json":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3.json","view_paper":"https://pith.science/paper/OX4YZRUK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.06058&json=true","fetch_graph":"https://pith.science/api/pith-number/OX4YZRUKD43NBJWMHZOHKADRC3/graph.json","fetch_events":"https://pith.science/api/pith-number/OX4YZRUKD43NBJWMHZOHKADRC3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3/action/storage_attestation","attest_author":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3/action/author_attestation","sign_citation":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3/action/citation_signature","submit_replication":"https://pith.science/pith/OX4YZRUKD43NBJWMHZOHKADRC3/action/replication_record"}},"created_at":"2026-07-05T09:38:49.133877+00:00","updated_at":"2026-07-05T09:38:49.133877+00:00"}