{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BD5ZO3FJ3RGJRIVK3QNKYEMYS2","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":"4637aba8d524e8ddce889383f55a7c2135d230c610305a5ecda90a2c28e7dc8f","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-27T19:28:43Z","title_canon_sha256":"7f84af1a9f6ac52965505dacb46cc063431634c7deae60d54f8233f3b030eb88"},"schema_version":"1.0","source":{"id":"2406.19501","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.19501","created_at":"2026-07-05T09:45:50Z"},{"alias_kind":"arxiv_version","alias_value":"2406.19501v2","created_at":"2026-07-05T09:45:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.19501","created_at":"2026-07-05T09:45:50Z"},{"alias_kind":"pith_short_12","alias_value":"BD5ZO3FJ3RGJ","created_at":"2026-07-05T09:45:50Z"},{"alias_kind":"pith_short_16","alias_value":"BD5ZO3FJ3RGJRIVK","created_at":"2026-07-05T09:45:50Z"},{"alias_kind":"pith_short_8","alias_value":"BD5ZO3FJ","created_at":"2026-07-05T09:45:50Z"}],"graph_snapshots":[{"event_id":"sha256:5a54277ab01fa83e62201e65538f4e67012be51475c1c53004c944cae482e710","target":"graph","created_at":"2026-07-05T09:45:50Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2406.19501/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models are susceptible to bias, sycophancy, backdoors, and other tendencies that lead to unfaithful responses to the input context. Interpreting internal states of language models could help monitor and correct unfaithful behavior. We hypothesize that language models represent their input contexts in a latent world model, and seek to extract this latent world state from the activations. We do so with 'propositional probes', which compositionally probe tokens for lexical information and bind them into logical propositions representing the world state. For example, given the input conte","authors_text":"Jacob Steinhardt, Jiahai Feng, Stuart Russell","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-27T19:28:43Z","title":"Monitoring Latent World States in Language Models with Propositional Probes"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.19501","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3dfd2eab9b974dcc9ac4c8a24894b21a2b483640280518d5e3c8487c8bf96388","target":"record","created_at":"2026-07-05T09:45:50Z","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":"4637aba8d524e8ddce889383f55a7c2135d230c610305a5ecda90a2c28e7dc8f","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-27T19:28:43Z","title_canon_sha256":"7f84af1a9f6ac52965505dacb46cc063431634c7deae60d54f8233f3b030eb88"},"schema_version":"1.0","source":{"id":"2406.19501","kind":"arxiv","version":2}},"canonical_sha256":"08fb976ca9dc4c98a2aadc1aac11989688279d0dee7467ffda23209c4731acf4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"08fb976ca9dc4c98a2aadc1aac11989688279d0dee7467ffda23209c4731acf4","first_computed_at":"2026-07-05T09:45:50.865097Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:45:50.865097Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YFljCrIjGibutKt/2XK0kLGGKzXYbjy+S6/QpfyhqRbNR6azJvnjhZgQESKr5LkHx9fiPJ7eC3Md7aNTxdZjDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:45:50.865554Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.19501","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3dfd2eab9b974dcc9ac4c8a24894b21a2b483640280518d5e3c8487c8bf96388","sha256:5a54277ab01fa83e62201e65538f4e67012be51475c1c53004c944cae482e710"],"state_sha256":"58ce3df85480a39b8c5635f75a547d0896134c5ccb54c43a6cbad7a579ad063a"}