{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:YX4Z6XPXK7OGKA2WDTVHOOWUY4","short_pith_number":"pith:YX4Z6XPX","schema_version":"1.0","canonical_sha256":"c5f99f5df757dc6503561cea773ad4c715b05fb7263d13d4d9c6cb1255bd1b6e","source":{"kind":"arxiv","id":"2403.15498","version":2},"attestation_state":"computed","paper":{"title":"Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Adam Karvonen","submitted_at":"2024-03-21T18:53:23Z","abstract_excerpt":"Language models have shown unprecedented capabilities, sparking debate over the source of their performance. Is it merely the outcome of learning syntactic patterns and surface level statistics, or do they extract semantics and a world model from the text? Prior work by Li et al. investigated this by training a GPT model on synthetic, randomly generated Othello games and found that the model learned an internal representation of the board state. We extend this work into the more complex domain of chess, training on real games and investigating our model's internal representations using linear "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2403.15498","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-21T18:53:23Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"3434c57cbe254643ed4e45fb76ed0d3ef49efbdc8ba83bc56b621651eaf778d7","abstract_canon_sha256":"0f0afca0c64d63e88c09b171eae55078151ef6a15c5aaa0752139f4485c84d50"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:43:27.352711Z","signature_b64":"lti6+H4Id5LmR5kGqjUQnIgPyiuSI8l73jFb09TKiV8O7nBAL2gIUpeyzhmpFchCK1U8W07F4YOuVZQNb/3CDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c5f99f5df757dc6503561cea773ad4c715b05fb7263d13d4d9c6cb1255bd1b6e","last_reissued_at":"2026-07-05T08:43:27.352229Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:43:27.352229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Adam Karvonen","submitted_at":"2024-03-21T18:53:23Z","abstract_excerpt":"Language models have shown unprecedented capabilities, sparking debate over the source of their performance. Is it merely the outcome of learning syntactic patterns and surface level statistics, or do they extract semantics and a world model from the text? Prior work by Li et al. investigated this by training a GPT model on synthetic, randomly generated Othello games and found that the model learned an internal representation of the board state. We extend this work into the more complex domain of chess, training on real games and investigating our model's internal representations using linear "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.15498","kind":"arxiv","version":2},"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/2403.15498/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2403.15498","created_at":"2026-07-05T08:43:27.352287+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.15498v2","created_at":"2026-07-05T08:43:27.352287+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.15498","created_at":"2026-07-05T08:43:27.352287+00:00"},{"alias_kind":"pith_short_12","alias_value":"YX4Z6XPXK7OG","created_at":"2026-07-05T08:43:27.352287+00:00"},{"alias_kind":"pith_short_16","alias_value":"YX4Z6XPXK7OGKA2W","created_at":"2026-07-05T08:43:27.352287+00:00"},{"alias_kind":"pith_short_8","alias_value":"YX4Z6XPX","created_at":"2026-07-05T08:43:27.352287+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03685","citing_title":"A Close Look At World Model Recovery In Supervised Fine-Tuned LLM Planners","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23972","citing_title":"Why We Need World Models for AGI: Where LLMs Fail and How World Models May Outperform","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4","json":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4.json","graph_json":"https://pith.science/api/pith-number/YX4Z6XPXK7OGKA2WDTVHOOWUY4/graph.json","events_json":"https://pith.science/api/pith-number/YX4Z6XPXK7OGKA2WDTVHOOWUY4/events.json","paper":"https://pith.science/paper/YX4Z6XPX"},"agent_actions":{"view_html":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4","download_json":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4.json","view_paper":"https://pith.science/paper/YX4Z6XPX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.15498&json=true","fetch_graph":"https://pith.science/api/pith-number/YX4Z6XPXK7OGKA2WDTVHOOWUY4/graph.json","fetch_events":"https://pith.science/api/pith-number/YX4Z6XPXK7OGKA2WDTVHOOWUY4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4/action/storage_attestation","attest_author":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4/action/author_attestation","sign_citation":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4/action/citation_signature","submit_replication":"https://pith.science/pith/YX4Z6XPXK7OGKA2WDTVHOOWUY4/action/replication_record"}},"created_at":"2026-07-05T08:43:27.352287+00:00","updated_at":"2026-07-05T08:43:27.352287+00:00"}