{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:DP2SXSCSVE2PZYJ27ZW7RCZK2I","short_pith_number":"pith:DP2SXSCS","canonical_record":{"source":{"id":"2510.22170","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-10-25T05:45:10Z","cross_cats_sorted":[],"title_canon_sha256":"2e3648f1a9798797790b145e87802116eed64f56e42beeb3158815142929d85b","abstract_canon_sha256":"57c76391ff39adb2728e1173ffc594c3c2d088bef551084767ac120cf583701a"},"schema_version":"1.0"},"canonical_sha256":"1bf52bc852a934fce13afe6df88b2ad23a6d96bc54357e2e07f21a99fb59e1b5","source":{"kind":"arxiv","id":"2510.22170","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.22170","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"arxiv_version","alias_value":"2510.22170v3","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.22170","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"pith_short_12","alias_value":"DP2SXSCSVE2P","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"pith_short_16","alias_value":"DP2SXSCSVE2PZYJ2","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"pith_short_8","alias_value":"DP2SXSCS","created_at":"2026-07-30T01:19:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:DP2SXSCSVE2PZYJ27ZW7RCZK2I","target":"record","payload":{"canonical_record":{"source":{"id":"2510.22170","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-10-25T05:45:10Z","cross_cats_sorted":[],"title_canon_sha256":"2e3648f1a9798797790b145e87802116eed64f56e42beeb3158815142929d85b","abstract_canon_sha256":"57c76391ff39adb2728e1173ffc594c3c2d088bef551084767ac120cf583701a"},"schema_version":"1.0"},"canonical_sha256":"1bf52bc852a934fce13afe6df88b2ad23a6d96bc54357e2e07f21a99fb59e1b5","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1bf52bc852a934fce13afe6df88b2ad23a6d96bc54357e2e07f21a99fb59e1b5","last_reissued_at":"2026-07-30T01:19:15.242007Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:19:15.242007Z"},"source_kind":"arxiv","source_id":"2510.22170","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-30T01:19:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NJOZV5t3+nfa0bsundDk0NboyTqKZwDJwDFvPVhq6sa4J2am1/BozlQ968OFrFytlDeq0FTSwWJPWNV1u1wEBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T02:09:41.581341Z"},"content_sha256":"9ab11f0eb0eca4ede35602f45ebd692119206dfafb7b969b753ef4c30f2b04ae","schema_version":"1.0","event_id":"sha256:9ab11f0eb0eca4ede35602f45ebd692119206dfafb7b969b753ef4c30f2b04ae"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:DP2SXSCSVE2PZYJ27ZW7RCZK2I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Measure what Matters: Psychometric Evaluation of AI with Situational Judgment Tests","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Situational judgment tests combined with multidimensional item response theory extract stable latent behavioral tendencies from persona-conditioned large language models.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alexandra Yost, Allen Roush, Amirali Abdullah, Grant Corser, Jacqueline Hammack, Nina Xu, Ravid Shwartz-Ziv, Shivam Raval, Shreyans Jain","submitted_at":"2025-10-25T05:45:10Z","abstract_excerpt":"Persona conditioning is widely used to steer large language model (LLM) behavior, but it is unclear whether it induces stable behavioral structure or superficial variation. We propose a framework to measure consistent behavioral tendencies using situational judgment tests (SJTs), multidimensional item response theory (MIRT), and structured synthetic personas, treating responses as observations of latent behavioral variables.\n  Across large-scale SJT and persona datasets, we find that persona-conditioned behaviors are stable across runs, latent trait scores predict external benchmarks (e.g., Tr"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Persona-conditioned behaviors are stable across runs, latent trait scores predict external benchmarks (e.g., TruthfulQA, EmoBench), and MIRT reveals consistent latent structure.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That responses to situational judgment tests can be treated as observations of stable latent behavioral variables in LLMs rather than prompt-dependent or superficial patterns.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Proposes SJTs and MIRT to measure consistent latent behavioral tendencies in LLMs, showing stability and predictive validity on external benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Situational judgment tests combined with multidimensional item response theory extract stable latent behavioral tendencies from persona-conditioned large language models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4b6bb8530b9e8dca1bc4011f60b4f45b99f7986e5e5005bde33fc575561e62c4"},"source":{"id":"2510.22170","kind":"arxiv","version":3},"verdict":{"id":"4ee7c384-e35e-4319-b714-ec5e106ce582","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-18T04:49:24.475383Z","strongest_claim":"Persona-conditioned behaviors are stable across runs, latent trait scores predict external benchmarks (e.g., TruthfulQA, EmoBench), and MIRT reveals consistent latent structure.","one_line_summary":"Proposes SJTs and MIRT to measure consistent latent behavioral tendencies in LLMs, showing stability and predictive validity on external benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That responses to situational judgment tests can be treated as observations of stable latent behavioral variables in LLMs rather than prompt-dependent or superficial patterns.","pith_extraction_headline":"Situational judgment tests combined with multidimensional item response theory extract stable latent behavioral tendencies from persona-conditioned large language models."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2510.22170/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":2,"snapshot_sha256":"7100164c901550f5249dab3e8b9ca674d373c70582bf7081f61cfdb60b1d9a2c"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"4ee7c384-e35e-4319-b714-ec5e106ce582"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-30T01:19:15Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+9rgWKCCUAjAw0bC2C2/Zg84wLMnADS1BtC/1wwV1QLAEsAMrkh+py1iRUHdueGdx/tFAEDzZTRxZsEoz97dCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T02:09:41.582065Z"},"content_sha256":"e1e36fbc707daf27408fdd7f79ea484105e5ef6d2bd095479a8375c5881e933b","schema_version":"1.0","event_id":"sha256:e1e36fbc707daf27408fdd7f79ea484105e5ef6d2bd095479a8375c5881e933b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DP2SXSCSVE2PZYJ27ZW7RCZK2I/bundle.json","state_url":"https://pith.science/pith/DP2SXSCSVE2PZYJ27ZW7RCZK2I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DP2SXSCSVE2PZYJ27ZW7RCZK2I/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":"2026-08-09T02:09:41Z","links":{"resolver":"https://pith.science/pith/DP2SXSCSVE2PZYJ27ZW7RCZK2I","bundle":"https://pith.science/pith/DP2SXSCSVE2PZYJ27ZW7RCZK2I/bundle.json","state":"https://pith.science/pith/DP2SXSCSVE2PZYJ27ZW7RCZK2I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DP2SXSCSVE2PZYJ27ZW7RCZK2I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:DP2SXSCSVE2PZYJ27ZW7RCZK2I","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":"57c76391ff39adb2728e1173ffc594c3c2d088bef551084767ac120cf583701a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-10-25T05:45:10Z","title_canon_sha256":"2e3648f1a9798797790b145e87802116eed64f56e42beeb3158815142929d85b"},"schema_version":"1.0","source":{"id":"2510.22170","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2510.22170","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"arxiv_version","alias_value":"2510.22170v3","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.22170","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"pith_short_12","alias_value":"DP2SXSCSVE2P","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"pith_short_16","alias_value":"DP2SXSCSVE2PZYJ2","created_at":"2026-07-30T01:19:15Z"},{"alias_kind":"pith_short_8","alias_value":"DP2SXSCS","created_at":"2026-07-30T01:19:15Z"}],"graph_snapshots":[{"event_id":"sha256:e1e36fbc707daf27408fdd7f79ea484105e5ef6d2bd095479a8375c5881e933b","target":"graph","created_at":"2026-07-30T01:19:15Z","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":"Persona-conditioned behaviors are stable across runs, latent trait scores predict external benchmarks (e.g., TruthfulQA, EmoBench), and MIRT reveals consistent latent structure."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That responses to situational judgment tests can be treated as observations of stable latent behavioral variables in LLMs rather than prompt-dependent or superficial patterns."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Proposes SJTs and MIRT to measure consistent latent behavioral tendencies in LLMs, showing stability and predictive validity on external benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Situational judgment tests combined with multidimensional item response theory extract stable latent behavioral tendencies from persona-conditioned large language models."}],"snapshot_sha256":"4b6bb8530b9e8dca1bc4011f60b4f45b99f7986e5e5005bde33fc575561e62c4"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"7100164c901550f5249dab3e8b9ca674d373c70582bf7081f61cfdb60b1d9a2c"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2510.22170/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Persona conditioning is widely used to steer large language model (LLM) behavior, but it is unclear whether it induces stable behavioral structure or superficial variation. We propose a framework to measure consistent behavioral tendencies using situational judgment tests (SJTs), multidimensional item response theory (MIRT), and structured synthetic personas, treating responses as observations of latent behavioral variables.\n  Across large-scale SJT and persona datasets, we find that persona-conditioned behaviors are stable across runs, latent trait scores predict external benchmarks (e.g., Tr","authors_text":"Alexandra Yost, Allen Roush, Amirali Abdullah, Grant Corser, Jacqueline Hammack, Nina Xu, Ravid Shwartz-Ziv, Shivam Raval, Shreyans Jain","cross_cats":[],"headline":"Situational judgment tests combined with multidimensional item response theory extract stable latent behavioral tendencies from persona-conditioned large language models.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-10-25T05:45:10Z","title":"Measure what Matters: Psychometric Evaluation of AI with Situational Judgment Tests"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.22170","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-18T04:49:24.475383Z","id":"4ee7c384-e35e-4319-b714-ec5e106ce582","model_set":{"reader":"grok-4.3"},"one_line_summary":"Proposes SJTs and MIRT to measure consistent latent behavioral tendencies in LLMs, showing stability and predictive validity on external benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Situational judgment tests combined with multidimensional item response theory extract stable latent behavioral tendencies from persona-conditioned large language models.","strongest_claim":"Persona-conditioned behaviors are stable across runs, latent trait scores predict external benchmarks (e.g., TruthfulQA, EmoBench), and MIRT reveals consistent latent structure.","weakest_assumption":"That responses to situational judgment tests can be treated as observations of stable latent behavioral variables in LLMs rather than prompt-dependent or superficial patterns."}},"verdict_id":"4ee7c384-e35e-4319-b714-ec5e106ce582"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9ab11f0eb0eca4ede35602f45ebd692119206dfafb7b969b753ef4c30f2b04ae","target":"record","created_at":"2026-07-30T01:19:15Z","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":"57c76391ff39adb2728e1173ffc594c3c2d088bef551084767ac120cf583701a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-10-25T05:45:10Z","title_canon_sha256":"2e3648f1a9798797790b145e87802116eed64f56e42beeb3158815142929d85b"},"schema_version":"1.0","source":{"id":"2510.22170","kind":"arxiv","version":3}},"canonical_sha256":"1bf52bc852a934fce13afe6df88b2ad23a6d96bc54357e2e07f21a99fb59e1b5","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1bf52bc852a934fce13afe6df88b2ad23a6d96bc54357e2e07f21a99fb59e1b5","first_computed_at":"2026-07-30T01:19:15.242007Z","kind":"pith_receipt","last_reissued_at":"2026-07-30T01:19:15.242007Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2510.22170","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9ab11f0eb0eca4ede35602f45ebd692119206dfafb7b969b753ef4c30f2b04ae","sha256:e1e36fbc707daf27408fdd7f79ea484105e5ef6d2bd095479a8375c5881e933b"],"state_sha256":"46f21506a41c137e17f1fc12145546d5d661086529d37cc5c51b3f5a0fc5599f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l70LOoPJt6V70qtt2Mgq6r0dnHl23d2qJ1+y5SP1do6r6p87Uk5awAoAcYPOl5Mnt98dhyjhJpOeRmSMkfgeAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T02:09:41.586450Z","bundle_sha256":"db57cd6e7b3addd2f8db17f359b776271edaf8ca5cc38b93c32e38384587ea48"}}