{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WDCLNMAR4UOHX3FIO2FBSXCWZH","short_pith_number":"pith:WDCLNMAR","schema_version":"1.0","canonical_sha256":"b0c4b6b011e51c7beca8768a195c56c9e134954516e64cb7fa13894f90ad7397","source":{"kind":"arxiv","id":"2308.01404","version":2},"attestation_state":"computed","paper":{"title":"Hoodwinked: Deception and Cooperation in a Text-Based Game for Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aidan O'Gara","submitted_at":"2023-07-05T17:22:09Z","abstract_excerpt":"Are current language models capable of deception and lie detection? We study this question by introducing a text-based game called $\\textit{Hoodwinked}$, inspired by Mafia and Among Us. Players are locked in a house and must find a key to escape, but one player is tasked with killing the others. Each time a murder is committed, the surviving players have a natural language discussion then vote to banish one player from the game. We conduct experiments with agents controlled by GPT-3, GPT-3.5, and GPT-4 and find evidence of deception and lie detection capabilities. The killer often denies their"},"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":"2308.01404","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-05T17:22:09Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"95d9252445e555cb73396095c101020ee172917d186fac59c6552c416305a465","abstract_canon_sha256":"83f6d5687d1411fbaa6096b405ce4cc14674e2fa2de111a8cfc0e4f7c624000f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:37:34.531221Z","signature_b64":"eW2IelYuownlihQoBOiUSQL2lW6CyPPnxCi8suOHWpmZZGpPdMKMAPf/CgSKmWhxzB9x0tulu9h9vW0rw/emBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0c4b6b011e51c7beca8768a195c56c9e134954516e64cb7fa13894f90ad7397","last_reissued_at":"2026-07-05T06:37:34.530754Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:37:34.530754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hoodwinked: Deception and Cooperation in a Text-Based Game for Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.CL","authors_text":"Aidan O'Gara","submitted_at":"2023-07-05T17:22:09Z","abstract_excerpt":"Are current language models capable of deception and lie detection? We study this question by introducing a text-based game called $\\textit{Hoodwinked}$, inspired by Mafia and Among Us. Players are locked in a house and must find a key to escape, but one player is tasked with killing the others. Each time a murder is committed, the surviving players have a natural language discussion then vote to banish one player from the game. We conduct experiments with agents controlled by GPT-3, GPT-3.5, and GPT-4 and find evidence of deception and lie detection capabilities. The killer often denies their"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.01404","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/2308.01404/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":"2308.01404","created_at":"2026-07-05T06:37:34.530807+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.01404v2","created_at":"2026-07-05T06:37:34.530807+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.01404","created_at":"2026-07-05T06:37:34.530807+00:00"},{"alias_kind":"pith_short_12","alias_value":"WDCLNMAR4UOH","created_at":"2026-07-05T06:37:34.530807+00:00"},{"alias_kind":"pith_short_16","alias_value":"WDCLNMAR4UOHX3FI","created_at":"2026-07-05T06:37:34.530807+00:00"},{"alias_kind":"pith_short_8","alias_value":"WDCLNMAR","created_at":"2026-07-05T06:37:34.530807+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31916","citing_title":"Theory of Mind and Persuasion Beyond Conversation: Assessing the Capacity of LLMs to Induce Belief States via Planning and Action","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2412.04984","citing_title":"Frontier Models are Capable of In-context Scheming","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16022","citing_title":"SocialGrid: A Benchmark for Planning and Social Reasoning in Embodied Multi-Agent Systems","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH","json":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH.json","graph_json":"https://pith.science/api/pith-number/WDCLNMAR4UOHX3FIO2FBSXCWZH/graph.json","events_json":"https://pith.science/api/pith-number/WDCLNMAR4UOHX3FIO2FBSXCWZH/events.json","paper":"https://pith.science/paper/WDCLNMAR"},"agent_actions":{"view_html":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH","download_json":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH.json","view_paper":"https://pith.science/paper/WDCLNMAR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.01404&json=true","fetch_graph":"https://pith.science/api/pith-number/WDCLNMAR4UOHX3FIO2FBSXCWZH/graph.json","fetch_events":"https://pith.science/api/pith-number/WDCLNMAR4UOHX3FIO2FBSXCWZH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH/action/storage_attestation","attest_author":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH/action/author_attestation","sign_citation":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH/action/citation_signature","submit_replication":"https://pith.science/pith/WDCLNMAR4UOHX3FIO2FBSXCWZH/action/replication_record"}},"created_at":"2026-07-05T06:37:34.530807+00:00","updated_at":"2026-07-05T06:37:34.530807+00:00"}