{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:HQQG2GM6ZLSNS4XM4A6VX3RPJA","short_pith_number":"pith:HQQG2GM6","canonical_record":{"source":{"id":"2406.03689","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T02:20:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"86091b5caef84f744808381126b1f6f4fa619c71edafcc30f6f130ea96a7cb42","abstract_canon_sha256":"d493feded3783febcaeb44854f68a8c16d4efb808d9323e2895c1d6a3bebdf58"},"schema_version":"1.0"},"canonical_sha256":"3c206d199ecae4d972ece03d5bee2f4814b1186e892c8a6bb927e8e39d2bb1ec","source":{"kind":"arxiv","id":"2406.03689","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.03689","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"arxiv_version","alias_value":"2406.03689v3","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.03689","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"pith_short_12","alias_value":"HQQG2GM6ZLSN","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"pith_short_16","alias_value":"HQQG2GM6ZLSNS4XM","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"pith_short_8","alias_value":"HQQG2GM6","created_at":"2026-07-05T09:33:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:HQQG2GM6ZLSNS4XM4A6VX3RPJA","target":"record","payload":{"canonical_record":{"source":{"id":"2406.03689","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T02:20:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"86091b5caef84f744808381126b1f6f4fa619c71edafcc30f6f130ea96a7cb42","abstract_canon_sha256":"d493feded3783febcaeb44854f68a8c16d4efb808d9323e2895c1d6a3bebdf58"},"schema_version":"1.0"},"canonical_sha256":"3c206d199ecae4d972ece03d5bee2f4814b1186e892c8a6bb927e8e39d2bb1ec","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:33:32.726600Z","signature_b64":"zd4zhfmcVEy2Focx/EBq/uN41+AktgXGUfANGPMrrSoIF3dNJtXNVpbP7WgDVUyjp4p7zaCctCgNMdyNcKTJBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3c206d199ecae4d972ece03d5bee2f4814b1186e892c8a6bb927e8e39d2bb1ec","last_reissued_at":"2026-07-05T09:33:32.726093Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:33:32.726093Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.03689","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":"2026-07-05T09:33:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+lzluxAdeiWTdWa4HktpisbHgp3Ucte+VDn5eGjR7lYInEagaeggJaVTZuCnIfa8E43264Q7JCL0as4i3Lh7AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:20:06.467332Z"},"content_sha256":"850291459d5edaee62f33f9babed5178a650ed40f80bfc2201a91401c627b037","schema_version":"1.0","event_id":"sha256:850291459d5edaee62f33f9babed5178a650ed40f80bfc2201a91401c627b037"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:HQQG2GM6ZLSNS4XM4A6VX3RPJA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Evaluating the World Model Implicit in a Generative Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ashesh Rambachan, Jon Kleinberg, Justin Y. Chen, Keyon Vafa, Sendhil Mullainathan","submitted_at":"2024-06-06T02:20:31Z","abstract_excerpt":"Recent work suggests that large language models may implicitly learn world models. How should we assess this possibility? We formalize this question for the case where the underlying reality is governed by a deterministic finite automaton. This includes problems as diverse as simple logical reasoning, geographic navigation, game-playing, and chemistry. We propose new evaluation metrics for world model recovery inspired by the classic Myhill-Nerode theorem from language theory. We illustrate their utility in three domains: game playing, logic puzzles, and navigation. In all domains, the generat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.03689","kind":"arxiv","version":3},"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/2406.03689/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:33:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PNJ6iFipUTNJ3PZNP5WKUPZJF9Cj+Z/vw46OeIkdp2nEMm+SqBG/qGgkNQWg8eIh96+LDnWmJi2R+uOM0MNKBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:20:06.468275Z"},"content_sha256":"bceea598f26e9b6fe788d896158ca674ba4bbd3ead1170fcc0e55c67b12efb88","schema_version":"1.0","event_id":"sha256:bceea598f26e9b6fe788d896158ca674ba4bbd3ead1170fcc0e55c67b12efb88"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HQQG2GM6ZLSNS4XM4A6VX3RPJA/bundle.json","state_url":"https://pith.science/pith/HQQG2GM6ZLSNS4XM4A6VX3RPJA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HQQG2GM6ZLSNS4XM4A6VX3RPJA/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":"2026-08-10T02:20:06Z","links":{"resolver":"https://pith.science/pith/HQQG2GM6ZLSNS4XM4A6VX3RPJA","bundle":"https://pith.science/pith/HQQG2GM6ZLSNS4XM4A6VX3RPJA/bundle.json","state":"https://pith.science/pith/HQQG2GM6ZLSNS4XM4A6VX3RPJA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HQQG2GM6ZLSNS4XM4A6VX3RPJA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:HQQG2GM6ZLSNS4XM4A6VX3RPJA","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":"d493feded3783febcaeb44854f68a8c16d4efb808d9323e2895c1d6a3bebdf58","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T02:20:31Z","title_canon_sha256":"86091b5caef84f744808381126b1f6f4fa619c71edafcc30f6f130ea96a7cb42"},"schema_version":"1.0","source":{"id":"2406.03689","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.03689","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"arxiv_version","alias_value":"2406.03689v3","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.03689","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"pith_short_12","alias_value":"HQQG2GM6ZLSN","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"pith_short_16","alias_value":"HQQG2GM6ZLSNS4XM","created_at":"2026-07-05T09:33:32Z"},{"alias_kind":"pith_short_8","alias_value":"HQQG2GM6","created_at":"2026-07-05T09:33:32Z"}],"graph_snapshots":[{"event_id":"sha256:bceea598f26e9b6fe788d896158ca674ba4bbd3ead1170fcc0e55c67b12efb88","target":"graph","created_at":"2026-07-05T09:33:32Z","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/2406.03689/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent work suggests that large language models may implicitly learn world models. How should we assess this possibility? We formalize this question for the case where the underlying reality is governed by a deterministic finite automaton. This includes problems as diverse as simple logical reasoning, geographic navigation, game-playing, and chemistry. We propose new evaluation metrics for world model recovery inspired by the classic Myhill-Nerode theorem from language theory. We illustrate their utility in three domains: game playing, logic puzzles, and navigation. In all domains, the generat","authors_text":"Ashesh Rambachan, Jon Kleinberg, Justin Y. Chen, Keyon Vafa, Sendhil Mullainathan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T02:20:31Z","title":"Evaluating the World Model Implicit in a Generative Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.03689","kind":"arxiv","version":3},"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":"sha256:850291459d5edaee62f33f9babed5178a650ed40f80bfc2201a91401c627b037","target":"record","created_at":"2026-07-05T09:33:32Z","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":"d493feded3783febcaeb44854f68a8c16d4efb808d9323e2895c1d6a3bebdf58","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T02:20:31Z","title_canon_sha256":"86091b5caef84f744808381126b1f6f4fa619c71edafcc30f6f130ea96a7cb42"},"schema_version":"1.0","source":{"id":"2406.03689","kind":"arxiv","version":3}},"canonical_sha256":"3c206d199ecae4d972ece03d5bee2f4814b1186e892c8a6bb927e8e39d2bb1ec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3c206d199ecae4d972ece03d5bee2f4814b1186e892c8a6bb927e8e39d2bb1ec","first_computed_at":"2026-07-05T09:33:32.726093Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:33:32.726093Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zd4zhfmcVEy2Focx/EBq/uN41+AktgXGUfANGPMrrSoIF3dNJtXNVpbP7WgDVUyjp4p7zaCctCgNMdyNcKTJBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:33:32.726600Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.03689","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:850291459d5edaee62f33f9babed5178a650ed40f80bfc2201a91401c627b037","sha256:bceea598f26e9b6fe788d896158ca674ba4bbd3ead1170fcc0e55c67b12efb88"],"state_sha256":"168e32e850b0ff5e9d64c9a19a4279c0d35431c51cc3f80214d4116f9af6eb78"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nh95MmBgP48EQrFcX4YZu4SWyEU0L/jnQRb6DV9Grz7NawQLKnAXyhhpJ62+34aEqQPH3cmstUdeYlVCeUdXBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T02:20:06.473774Z","bundle_sha256":"74ef859a265b2e591cacd697d5faebd3e12eff1f89bd0abbc9f55ba079fbda87"}}