{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:RKO7NA5EKQANGMFQBPVY7QNLI4","short_pith_number":"pith:RKO7NA5E","canonical_record":{"source":{"id":"2508.19851","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-27T13:08:51Z","cross_cats_sorted":[],"title_canon_sha256":"a2a1716df400475f9f5f9bd7e117a015cb2720ee257c5b97694501f0fdaac5bf","abstract_canon_sha256":"eb00d53fed86053da168c64104c409bd133a515122586b11205346e37a668f39"},"schema_version":"1.0"},"canonical_sha256":"8a9df683a45400d330b00beb8fc1ab4723e2d3638a3a0c8625771abe22516bdc","source":{"kind":"arxiv","id":"2508.19851","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.19851","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"arxiv_version","alias_value":"2508.19851v1","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19851","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"pith_short_12","alias_value":"RKO7NA5EKQAN","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"pith_short_16","alias_value":"RKO7NA5EKQANGMFQ","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"pith_short_8","alias_value":"RKO7NA5E","created_at":"2026-07-05T12:00:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:RKO7NA5EKQANGMFQBPVY7QNLI4","target":"record","payload":{"canonical_record":{"source":{"id":"2508.19851","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-27T13:08:51Z","cross_cats_sorted":[],"title_canon_sha256":"a2a1716df400475f9f5f9bd7e117a015cb2720ee257c5b97694501f0fdaac5bf","abstract_canon_sha256":"eb00d53fed86053da168c64104c409bd133a515122586b11205346e37a668f39"},"schema_version":"1.0"},"canonical_sha256":"8a9df683a45400d330b00beb8fc1ab4723e2d3638a3a0c8625771abe22516bdc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:00:18.740309Z","signature_b64":"gEa1aev8nbVpMhwkPCoMB9XRsXtRESJ/t59e1qsFZhCeSWkzeDi86j4EITZFTKKHTJZ1lnfAZ8ZM0zy3D81/Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a9df683a45400d330b00beb8fc1ab4723e2d3638a3a0c8625771abe22516bdc","last_reissued_at":"2026-07-05T12:00:18.739806Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:00:18.739806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.19851","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:00:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gtXZ0E8IkNtwo8CRXQDtvJGPte4jUVIT4ELDGzOySdc2rGnc62ZaeGELNnJFXNmu8Q01bG4f9wF5DIZdVW9uAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T07:26:26.009820Z"},"content_sha256":"262c5a9c717e7e85783e9521959c7237fab18ce153ecbd33184a0c49423770bc","schema_version":"1.0","event_id":"sha256:262c5a9c717e7e85783e9521959c7237fab18ce153ecbd33184a0c49423770bc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:RKO7NA5EKQANGMFQBPVY7QNLI4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tracking World States with Language Models: State-Based Evaluation Using Chess","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Jason Naradowsky, Romain Harang, Yaswitha Gujju, Yusuke Miyao","submitted_at":"2025-08-27T13:08:51Z","abstract_excerpt":"Large Language Models (LLMs) exhibit emergent capabilities in structured domains, suggesting they may implicitly internalize high-fidelity representations of world models. While probing techniques have shown promising signs of this in scientific and game-based settings, they rely on model-specific internal activations, which limit interpretability and generalizability. In this work, we propose a model-agnostic, state-based evaluation framework using chess as a benchmark to assess whether LLMs preserve the semantics of structured environments. Our method analyzes the downstream legal move distr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19851","kind":"arxiv","version":1},"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/2508.19851/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":"2026-07-05T12:00:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"R5juSygrx11dhLPJPDmJaSVforGHJbkC7bUHmfDfXCm09CxADwwu3SnmV9Cq1H8OGInfKYQYRmfyarfTPpPnCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-23T07:26:26.010345Z"},"content_sha256":"f6b233bba112293647d07bfffe6f8d1536019f068105f91d16f88213cb81e396","schema_version":"1.0","event_id":"sha256:f6b233bba112293647d07bfffe6f8d1536019f068105f91d16f88213cb81e396"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RKO7NA5EKQANGMFQBPVY7QNLI4/bundle.json","state_url":"https://pith.science/pith/RKO7NA5EKQANGMFQBPVY7QNLI4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RKO7NA5EKQANGMFQBPVY7QNLI4/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":"2026-08-23T07:26:26Z","links":{"resolver":"https://pith.science/pith/RKO7NA5EKQANGMFQBPVY7QNLI4","bundle":"https://pith.science/pith/RKO7NA5EKQANGMFQBPVY7QNLI4/bundle.json","state":"https://pith.science/pith/RKO7NA5EKQANGMFQBPVY7QNLI4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RKO7NA5EKQANGMFQBPVY7QNLI4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:RKO7NA5EKQANGMFQBPVY7QNLI4","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":"eb00d53fed86053da168c64104c409bd133a515122586b11205346e37a668f39","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-27T13:08:51Z","title_canon_sha256":"a2a1716df400475f9f5f9bd7e117a015cb2720ee257c5b97694501f0fdaac5bf"},"schema_version":"1.0","source":{"id":"2508.19851","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.19851","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"arxiv_version","alias_value":"2508.19851v1","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19851","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"pith_short_12","alias_value":"RKO7NA5EKQAN","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"pith_short_16","alias_value":"RKO7NA5EKQANGMFQ","created_at":"2026-07-05T12:00:18Z"},{"alias_kind":"pith_short_8","alias_value":"RKO7NA5E","created_at":"2026-07-05T12:00:18Z"}],"graph_snapshots":[{"event_id":"sha256:f6b233bba112293647d07bfffe6f8d1536019f068105f91d16f88213cb81e396","target":"graph","created_at":"2026-07-05T12:00:18Z","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/2508.19851/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) exhibit emergent capabilities in structured domains, suggesting they may implicitly internalize high-fidelity representations of world models. While probing techniques have shown promising signs of this in scientific and game-based settings, they rely on model-specific internal activations, which limit interpretability and generalizability. In this work, we propose a model-agnostic, state-based evaluation framework using chess as a benchmark to assess whether LLMs preserve the semantics of structured environments. Our method analyzes the downstream legal move distr","authors_text":"Jason Naradowsky, Romain Harang, Yaswitha Gujju, Yusuke Miyao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-27T13:08:51Z","title":"Tracking World States with Language Models: State-Based Evaluation Using Chess"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19851","kind":"arxiv","version":1},"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":"sha256:262c5a9c717e7e85783e9521959c7237fab18ce153ecbd33184a0c49423770bc","target":"record","created_at":"2026-07-05T12:00:18Z","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":"eb00d53fed86053da168c64104c409bd133a515122586b11205346e37a668f39","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-27T13:08:51Z","title_canon_sha256":"a2a1716df400475f9f5f9bd7e117a015cb2720ee257c5b97694501f0fdaac5bf"},"schema_version":"1.0","source":{"id":"2508.19851","kind":"arxiv","version":1}},"canonical_sha256":"8a9df683a45400d330b00beb8fc1ab4723e2d3638a3a0c8625771abe22516bdc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8a9df683a45400d330b00beb8fc1ab4723e2d3638a3a0c8625771abe22516bdc","first_computed_at":"2026-07-05T12:00:18.739806Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:00:18.739806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gEa1aev8nbVpMhwkPCoMB9XRsXtRESJ/t59e1qsFZhCeSWkzeDi86j4EITZFTKKHTJZ1lnfAZ8ZM0zy3D81/Ag==","signature_status":"signed_v1","signed_at":"2026-07-05T12:00:18.740309Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.19851","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:262c5a9c717e7e85783e9521959c7237fab18ce153ecbd33184a0c49423770bc","sha256:f6b233bba112293647d07bfffe6f8d1536019f068105f91d16f88213cb81e396"],"state_sha256":"f8820357a7498319df7a6c26d89c9c22bf9764a434d0c4fe33cdbfcf3f6677a1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"o9aX17VDskAll0iR/tEg1OH+5NGDjnSa2JzCzkw1hMB3rr9q9DFO/NyUhz5fgEOBNywcy2PrrCm3haIK8JMWDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-23T07:26:26.014813Z","bundle_sha256":"591d3fbd1cdce5d0579d015ea32a8e3e829b4a92ee613330aa6174f82c951c85"}}