{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:D55O2SY2TCSB5H2XUIHELH2W5S","short_pith_number":"pith:D55O2SY2","schema_version":"1.0","canonical_sha256":"1f7aed4b1a98a41e9f57a20e459f56eca47ed3adf8a379fe2949d101272f6dfb","source":{"kind":"arxiv","id":"2502.01591","version":3},"attestation_state":"computed","paper":{"title":"Improving Transformer World Models for Data-Efficient RL","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Antoine Dedieu, Carter Wendelken, Joseph Ortiz, J Swaroop Guntupalli, Kevin Patrick Murphy, Miguel Lazaro-Gredilla, Wolfgang Lehrach, Xinghua Lou","submitted_at":"2025-02-03T18:25:17Z","abstract_excerpt":"We present three improvements to the standard model-based RL paradigm based on transformers: (a) \"Dyna with warmup\", which trains the policy on real and imaginary data, but only starts using imaginary data after the world model has been sufficiently trained; (b) \"nearest neighbor tokenizer\" for image patches, which improves upon previous tokenization schemes, which are needed when using a transformer world model (TWM), by ensuring the code words are static after creation, thus providing a constant target for TWM learning; and (c) \"block teacher forcing\", which allows the TWM to reason jointly "},"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":"2502.01591","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-03T18:25:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3eb643c6661d2939d61a50c71cd83fd76afaa0b5f0f62b9698102235740134aa","abstract_canon_sha256":"f44383a4e7225ab1ef3be518bb1343c56cdbd90da025a17e7491cfbbcec3c24d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:38:26.750416Z","signature_b64":"OuIDudP6JuHHnpNI0VlgMaea57a62vjSKL+sMKhZoISSdglBYwRjf/rDux20rVkZ+olqT4YkcOMok8uBkuw2AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1f7aed4b1a98a41e9f57a20e459f56eca47ed3adf8a379fe2949d101272f6dfb","last_reissued_at":"2026-07-05T11:38:26.749849Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:38:26.749849Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Transformer World Models for Data-Efficient RL","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Antoine Dedieu, Carter Wendelken, Joseph Ortiz, J Swaroop Guntupalli, Kevin Patrick Murphy, Miguel Lazaro-Gredilla, Wolfgang Lehrach, Xinghua Lou","submitted_at":"2025-02-03T18:25:17Z","abstract_excerpt":"We present three improvements to the standard model-based RL paradigm based on transformers: (a) \"Dyna with warmup\", which trains the policy on real and imaginary data, but only starts using imaginary data after the world model has been sufficiently trained; (b) \"nearest neighbor tokenizer\" for image patches, which improves upon previous tokenization schemes, which are needed when using a transformer world model (TWM), by ensuring the code words are static after creation, thus providing a constant target for TWM learning; and (c) \"block teacher forcing\", which allows the TWM to reason jointly "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.01591","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/2502.01591/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":"2502.01591","created_at":"2026-07-05T11:38:26.749911+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.01591v3","created_at":"2026-07-05T11:38:26.749911+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.01591","created_at":"2026-07-05T11:38:26.749911+00:00"},{"alias_kind":"pith_short_12","alias_value":"D55O2SY2TCSB","created_at":"2026-07-05T11:38:26.749911+00:00"},{"alias_kind":"pith_short_16","alias_value":"D55O2SY2TCSB5H2X","created_at":"2026-07-05T11:38:26.749911+00:00"},{"alias_kind":"pith_short_8","alias_value":"D55O2SY2","created_at":"2026-07-05T11:38:26.749911+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00811","citing_title":"From Pixels to Temporal Correlations: Learning Informative Representations for Reinforcement Learning Pre-training","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2502.11537","citing_title":"Simulus: Combining Improvements in Sample-Efficient World Model Agents","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S","json":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S.json","graph_json":"https://pith.science/api/pith-number/D55O2SY2TCSB5H2XUIHELH2W5S/graph.json","events_json":"https://pith.science/api/pith-number/D55O2SY2TCSB5H2XUIHELH2W5S/events.json","paper":"https://pith.science/paper/D55O2SY2"},"agent_actions":{"view_html":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S","download_json":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S.json","view_paper":"https://pith.science/paper/D55O2SY2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.01591&json=true","fetch_graph":"https://pith.science/api/pith-number/D55O2SY2TCSB5H2XUIHELH2W5S/graph.json","fetch_events":"https://pith.science/api/pith-number/D55O2SY2TCSB5H2XUIHELH2W5S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S/action/storage_attestation","attest_author":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S/action/author_attestation","sign_citation":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S/action/citation_signature","submit_replication":"https://pith.science/pith/D55O2SY2TCSB5H2XUIHELH2W5S/action/replication_record"}},"created_at":"2026-07-05T11:38:26.749911+00:00","updated_at":"2026-07-05T11:38:26.749911+00:00"}