{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4SUPBB6H5VM6CIDKVCA3FIB4YB","short_pith_number":"pith:4SUPBB6H","schema_version":"1.0","canonical_sha256":"e4a8f087c7ed59e1206aa881b2a03cc05848bfb769f648302d88c2c5a5f96c5a","source":{"kind":"arxiv","id":"2507.02197","version":1},"attestation_state":"computed","paper":{"title":"Do Role-Playing Agents Practice What They Preach? Belief-Behavior Consistency in LLM-Based Simulations of Human Trust","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Adam Davies, Amogh Mannekote, Bonnie J Dorr, ChengXiang Zhai, Francesco Pinto, Guohao Li, Kristy Elizabeth Boyer","submitted_at":"2025-07-02T23:30:51Z","abstract_excerpt":"As LLMs are increasingly studied as role-playing agents to generate synthetic data for human behavioral research, ensuring that their outputs remain coherent with their assigned roles has become a critical concern. In this paper, we investigate how consistently LLM-based role-playing agents' stated beliefs about the behavior of the people they are asked to role-play (\"what they say\") correspond to their actual behavior during role-play (\"how they act\"). Specifically, we establish an evaluation framework to rigorously measure how well beliefs obtained by prompting the model can predict simulati"},"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":"2507.02197","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-07-02T23:30:51Z","cross_cats_sorted":[],"title_canon_sha256":"d430689b098f8781e7bdfa095f5e29d8d427ce873295c05b622af1744b32c8c7","abstract_canon_sha256":"d3f59ef3d92175e06c1c906591c5945599515b80bc521dcecd7e40406021a4c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:31:13.640065Z","signature_b64":"2nASU0pTpXZb9x1Ji7RD278huRIeJkBbg5Kf62dJSrgo4dlrRKCD+7m0/6JbXNlB3sb3+eJbTU9ayoxk57wjCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4a8f087c7ed59e1206aa881b2a03cc05848bfb769f648302d88c2c5a5f96c5a","last_reissued_at":"2026-07-05T11:31:13.639398Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:31:13.639398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Role-Playing Agents Practice What They Preach? Belief-Behavior Consistency in LLM-Based Simulations of Human Trust","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Adam Davies, Amogh Mannekote, Bonnie J Dorr, ChengXiang Zhai, Francesco Pinto, Guohao Li, Kristy Elizabeth Boyer","submitted_at":"2025-07-02T23:30:51Z","abstract_excerpt":"As LLMs are increasingly studied as role-playing agents to generate synthetic data for human behavioral research, ensuring that their outputs remain coherent with their assigned roles has become a critical concern. In this paper, we investigate how consistently LLM-based role-playing agents' stated beliefs about the behavior of the people they are asked to role-play (\"what they say\") correspond to their actual behavior during role-play (\"how they act\"). Specifically, we establish an evaluation framework to rigorously measure how well beliefs obtained by prompting the model can predict simulati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.02197","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/2507.02197/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":"2507.02197","created_at":"2026-07-05T11:31:13.639523+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.02197v1","created_at":"2026-07-05T11:31:13.639523+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.02197","created_at":"2026-07-05T11:31:13.639523+00:00"},{"alias_kind":"pith_short_12","alias_value":"4SUPBB6H5VM6","created_at":"2026-07-05T11:31:13.639523+00:00"},{"alias_kind":"pith_short_16","alias_value":"4SUPBB6H5VM6CIDK","created_at":"2026-07-05T11:31:13.639523+00:00"},{"alias_kind":"pith_short_8","alias_value":"4SUPBB6H","created_at":"2026-07-05T11:31:13.639523+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30930","citing_title":"TUX: Measuring Human--AI Tacit Understanding","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB","json":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB.json","graph_json":"https://pith.science/api/pith-number/4SUPBB6H5VM6CIDKVCA3FIB4YB/graph.json","events_json":"https://pith.science/api/pith-number/4SUPBB6H5VM6CIDKVCA3FIB4YB/events.json","paper":"https://pith.science/paper/4SUPBB6H"},"agent_actions":{"view_html":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB","download_json":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB.json","view_paper":"https://pith.science/paper/4SUPBB6H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.02197&json=true","fetch_graph":"https://pith.science/api/pith-number/4SUPBB6H5VM6CIDKVCA3FIB4YB/graph.json","fetch_events":"https://pith.science/api/pith-number/4SUPBB6H5VM6CIDKVCA3FIB4YB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB/action/storage_attestation","attest_author":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB/action/author_attestation","sign_citation":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB/action/citation_signature","submit_replication":"https://pith.science/pith/4SUPBB6H5VM6CIDKVCA3FIB4YB/action/replication_record"}},"created_at":"2026-07-05T11:31:13.639523+00:00","updated_at":"2026-07-05T11:31:13.639523+00:00"}