{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AO7CP7YRR6UWOBKLXHMPQ4SXIG","short_pith_number":"pith:AO7CP7YR","canonical_record":{"source":{"id":"2509.10078","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-12T09:14:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"927259d1703836f385d2729fa5d0fef9560c5311460c0f82c681fc7375958df1","abstract_canon_sha256":"09be603bfdee750cb5415e56108a7c7d6e9a1f8991c91343038a881743e93d80"},"schema_version":"1.0"},"canonical_sha256":"03be27ff118fa967054bb9d8f8725741837f2718c8d515ec63047df6ae320b2d","source":{"kind":"arxiv","id":"2509.10078","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.10078","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"arxiv_version","alias_value":"2509.10078v4","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.10078","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"pith_short_12","alias_value":"AO7CP7YRR6UW","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"pith_short_16","alias_value":"AO7CP7YRR6UWOBKL","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"pith_short_8","alias_value":"AO7CP7YR","created_at":"2026-06-01T02:03:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AO7CP7YRR6UWOBKLXHMPQ4SXIG","target":"record","payload":{"canonical_record":{"source":{"id":"2509.10078","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-12T09:14:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"927259d1703836f385d2729fa5d0fef9560c5311460c0f82c681fc7375958df1","abstract_canon_sha256":"09be603bfdee750cb5415e56108a7c7d6e9a1f8991c91343038a881743e93d80"},"schema_version":"1.0"},"canonical_sha256":"03be27ff118fa967054bb9d8f8725741837f2718c8d515ec63047df6ae320b2d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T02:03:26.333497Z","signature_b64":"HzXiwRpIkl7Y+/4IsWujUSll0BuGb9UpC5dQhcNL/vSQb75bhu72qKZPnXIJzKIYgKVxZQUB+v1svWHpk2HFBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03be27ff118fa967054bb9d8f8725741837f2718c8d515ec63047df6ae320b2d","last_reissued_at":"2026-06-01T02:03:26.332380Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T02:03:26.332380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.10078","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-01T02:03:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bh0v+23lCKmkIlrRwH8p5jJ3rCtiAKg09CRpOtzWahFA2wE2VnGTuDHeIxLYpIpfaSptbKf8jMPvczeuQq0nDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T10:46:44.608045Z"},"content_sha256":"5c5e4d7a31089a175bcf080e41a434bd0fd2d156572509d138720522e35fe831","schema_version":"1.0","event_id":"sha256:5c5e4d7a31089a175bcf080e41a434bd0fd2d156572509d138720522e35fe831"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AO7CP7YRR6UWOBKLXHMPQ4SXIG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Human Psychometric Questionnaires Mischaracterize LLM Behavior","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"LLM responses to human psychometric questionnaires substantially differ from their generation probabilities on real-world user queries.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dongmin Choi, Eun-Ju Lee, Jongwook Han, Woojung Song, Yohan Jo, Yoonah Park","submitted_at":"2025-09-12T09:14:42Z","abstract_excerpt":"We examine whether human psychometric questionnaires can serve as reliable tools for characterizing and predicting LLM behavior in everyday user interactions. We analyze eight open-source LLMs by comparing their value and personality profiles derived from two different methods: Likert self-reports on established questionnaires (PVQ-40/21 and BFI-44/10) and generation probabilities over value-laden responses to everyday user queries. The two profiles diverge substantially. Within-construct item consistency, often cited as evidence of stable LLM dispositions, disappears in generation probabiliti"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The two profiles turn out to be substantially different and provide evidence that LLMs' responses to established questionnaires reflect desired behavior rather than stable psychological constructs, which challenges the consistent psychological dispositions of LLMs claimed in prior work.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That generation probability scores of value- or personality-laden responses to real-world user queries accurately capture the LLMs' psychological characteristics expressed during interactions with users.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Questionnaire-based and generation-based psychological profiles for LLMs are substantially different, indicating that established human questionnaires reflect desired behavior instead of stable psychological constructs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM responses to human psychometric questionnaires substantially differ from their generation probabilities on real-world user queries.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9c177c8f3bf3633f162f262a889c75e1357db9d289d2697beada8c05d53e7e21"},"source":{"id":"2509.10078","kind":"arxiv","version":4},"verdict":{"id":"02a33158-3c04-400c-ac23-fcf2243ef90a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-18T17:39:22.412131Z","strongest_claim":"The two profiles turn out to be substantially different and provide evidence that LLMs' responses to established questionnaires reflect desired behavior rather than stable psychological constructs, which challenges the consistent psychological dispositions of LLMs claimed in prior work.","one_line_summary":"Questionnaire-based and generation-based psychological profiles for LLMs are substantially different, indicating that established human questionnaires reflect desired behavior instead of stable psychological constructs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That generation probability scores of value- or personality-laden responses to real-world user queries accurately capture the LLMs' psychological characteristics expressed during interactions with users.","pith_extraction_headline":"LLM responses to human psychometric questionnaires substantially differ from their generation probabilities on real-world user queries."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2509.10078/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":1,"snapshot_sha256":"22d48b4e3d62abed5f4535c32699069d5658c7e213125fc275fbba2004467f8e"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"02a33158-3c04-400c-ac23-fcf2243ef90a"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-01T02:03:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ppvheRHx5H5B/cu2V4YP9SzE/hqQfw1ui20dDRMqEGhpHmgy3hrBHthpauVt0m4dznic/vyxvH2b6fo3SlPuCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T10:46:44.608523Z"},"content_sha256":"5cace9405ea827691983ee22f63c1302d474b8ede5fcdd96299493d92c1f58d4","schema_version":"1.0","event_id":"sha256:5cace9405ea827691983ee22f63c1302d474b8ede5fcdd96299493d92c1f58d4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AO7CP7YRR6UWOBKLXHMPQ4SXIG/bundle.json","state_url":"https://pith.science/pith/AO7CP7YRR6UWOBKLXHMPQ4SXIG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AO7CP7YRR6UWOBKLXHMPQ4SXIG/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-08T10:46:44Z","links":{"resolver":"https://pith.science/pith/AO7CP7YRR6UWOBKLXHMPQ4SXIG","bundle":"https://pith.science/pith/AO7CP7YRR6UWOBKLXHMPQ4SXIG/bundle.json","state":"https://pith.science/pith/AO7CP7YRR6UWOBKLXHMPQ4SXIG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AO7CP7YRR6UWOBKLXHMPQ4SXIG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AO7CP7YRR6UWOBKLXHMPQ4SXIG","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"09be603bfdee750cb5415e56108a7c7d6e9a1f8991c91343038a881743e93d80","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-12T09:14:42Z","title_canon_sha256":"927259d1703836f385d2729fa5d0fef9560c5311460c0f82c681fc7375958df1"},"schema_version":"1.0","source":{"id":"2509.10078","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.10078","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"arxiv_version","alias_value":"2509.10078v4","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.10078","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"pith_short_12","alias_value":"AO7CP7YRR6UW","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"pith_short_16","alias_value":"AO7CP7YRR6UWOBKL","created_at":"2026-06-01T02:03:26Z"},{"alias_kind":"pith_short_8","alias_value":"AO7CP7YR","created_at":"2026-06-01T02:03:26Z"}],"graph_snapshots":[{"event_id":"sha256:5cace9405ea827691983ee22f63c1302d474b8ede5fcdd96299493d92c1f58d4","target":"graph","created_at":"2026-06-01T02:03:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"The two profiles turn out to be substantially different and provide evidence that LLMs' responses to established questionnaires reflect desired behavior rather than stable psychological constructs, which challenges the consistent psychological dispositions of LLMs claimed in prior work."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That generation probability scores of value- or personality-laden responses to real-world user queries accurately capture the LLMs' psychological characteristics expressed during interactions with users."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Questionnaire-based and generation-based psychological profiles for LLMs are substantially different, indicating that established human questionnaires reflect desired behavior instead of stable psychological constructs."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLM responses to human psychometric questionnaires substantially differ from their generation probabilities on real-world user queries."}],"snapshot_sha256":"9c177c8f3bf3633f162f262a889c75e1357db9d289d2697beada8c05d53e7e21"},"formal_canon":{"evidence_count":1,"snapshot_sha256":"22d48b4e3d62abed5f4535c32699069d5658c7e213125fc275fbba2004467f8e"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2509.10078/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We examine whether human psychometric questionnaires can serve as reliable tools for characterizing and predicting LLM behavior in everyday user interactions. We analyze eight open-source LLMs by comparing their value and personality profiles derived from two different methods: Likert self-reports on established questionnaires (PVQ-40/21 and BFI-44/10) and generation probabilities over value-laden responses to everyday user queries. The two profiles diverge substantially. Within-construct item consistency, often cited as evidence of stable LLM dispositions, disappears in generation probabiliti","authors_text":"Dongmin Choi, Eun-Ju Lee, Jongwook Han, Woojung Song, Yohan Jo, Yoonah Park","cross_cats":["cs.AI"],"headline":"LLM responses to human psychometric questionnaires substantially differ from their generation probabilities on real-world user queries.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-12T09:14:42Z","title":"Human Psychometric Questionnaires Mischaracterize LLM Behavior"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.10078","kind":"arxiv","version":4},"verdict":{"created_at":"2026-05-18T17:39:22.412131Z","id":"02a33158-3c04-400c-ac23-fcf2243ef90a","model_set":{"reader":"grok-4.3"},"one_line_summary":"Questionnaire-based and generation-based psychological profiles for LLMs are substantially different, indicating that established human questionnaires reflect desired behavior instead of stable psychological constructs.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"LLM responses to human psychometric questionnaires substantially differ from their generation probabilities on real-world user queries.","strongest_claim":"The two profiles turn out to be substantially different and provide evidence that LLMs' responses to established questionnaires reflect desired behavior rather than stable psychological constructs, which challenges the consistent psychological dispositions of LLMs claimed in prior work.","weakest_assumption":"That generation probability scores of value- or personality-laden responses to real-world user queries accurately capture the LLMs' psychological characteristics expressed during interactions with users."}},"verdict_id":"02a33158-3c04-400c-ac23-fcf2243ef90a"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5c5e4d7a31089a175bcf080e41a434bd0fd2d156572509d138720522e35fe831","target":"record","created_at":"2026-06-01T02:03:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"09be603bfdee750cb5415e56108a7c7d6e9a1f8991c91343038a881743e93d80","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-09-12T09:14:42Z","title_canon_sha256":"927259d1703836f385d2729fa5d0fef9560c5311460c0f82c681fc7375958df1"},"schema_version":"1.0","source":{"id":"2509.10078","kind":"arxiv","version":4}},"canonical_sha256":"03be27ff118fa967054bb9d8f8725741837f2718c8d515ec63047df6ae320b2d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"03be27ff118fa967054bb9d8f8725741837f2718c8d515ec63047df6ae320b2d","first_computed_at":"2026-06-01T02:03:26.332380Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-01T02:03:26.332380Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HzXiwRpIkl7Y+/4IsWujUSll0BuGb9UpC5dQhcNL/vSQb75bhu72qKZPnXIJzKIYgKVxZQUB+v1svWHpk2HFBQ==","signature_status":"signed_v1","signed_at":"2026-06-01T02:03:26.333497Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.10078","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5c5e4d7a31089a175bcf080e41a434bd0fd2d156572509d138720522e35fe831","sha256:5cace9405ea827691983ee22f63c1302d474b8ede5fcdd96299493d92c1f58d4"],"state_sha256":"596b5214846ea93b49f2ed27a136ca3fa118773e49d4410c41ff9397b7864e6e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YJ4PY76YHl6u5cu9dSTJVrb64RAcP4H/W0Mj0Cg1pEmll2lUxgOORAa3rTKFvftAeuwNvN87YqfDy3cIBKXCBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-08T10:46:44.614115Z","bundle_sha256":"7118e1d3db4e592a347829153805549c6d9edf6fde576286fde61f29a857643f"}}