{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3CXSQH3WLKBPMDPNEWOHSJPR4S","short_pith_number":"pith:3CXSQH3W","schema_version":"1.0","canonical_sha256":"d8af281f765a82f60ded259c7925f1e4a3eb4d356aa720ed330cb0522b14884d","source":{"kind":"arxiv","id":"2406.01361","version":1},"attestation_state":"computed","paper":{"title":"Learning to Play Atari in a World of Tokens","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Pranav Agarwal, Samira Ebrahimi Kahou, Sheldon Andrews","submitted_at":"2024-06-03T14:25:29Z","abstract_excerpt":"Model-based reinforcement learning agents utilizing transformers have shown improved sample efficiency due to their ability to model extended context, resulting in more accurate world models. However, for complex reasoning and planning tasks, these methods primarily rely on continuous representations. This complicates modeling of discrete properties of the real world such as disjoint object classes between which interpolation is not plausible. In this work, we introduce discrete abstract representations for transformer-based learning (DART), a sample-efficient method utilizing discrete represe"},"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":"2406.01361","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-03T14:25:29Z","cross_cats_sorted":[],"title_canon_sha256":"02a09dfa50ff9f76d0090164e3315c9849f1f6fed2a89ba98f8c0a0ba7c0ded6","abstract_canon_sha256":"318a871c6db1799f1a2ce810b4f618c04b593835a0f9c8f41456c85412dc1ba4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:38.719559Z","signature_b64":"h/RF/PtlCyGCWbSmIUBjoEW5B1jlvSnKMnI68TFnNS0QT2l193r4XYyuT+vd8t7avSUdXTYj+YNsl4UXgYvbBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8af281f765a82f60ded259c7925f1e4a3eb4d356aa720ed330cb0522b14884d","last_reissued_at":"2026-07-05T08:26:38.719159Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:38.719159Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Play Atari in a World of Tokens","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Pranav Agarwal, Samira Ebrahimi Kahou, Sheldon Andrews","submitted_at":"2024-06-03T14:25:29Z","abstract_excerpt":"Model-based reinforcement learning agents utilizing transformers have shown improved sample efficiency due to their ability to model extended context, resulting in more accurate world models. However, for complex reasoning and planning tasks, these methods primarily rely on continuous representations. This complicates modeling of discrete properties of the real world such as disjoint object classes between which interpolation is not plausible. In this work, we introduce discrete abstract representations for transformer-based learning (DART), a sample-efficient method utilizing discrete represe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.01361","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/2406.01361/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":"2406.01361","created_at":"2026-07-05T08:26:38.719222+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.01361v1","created_at":"2026-07-05T08:26:38.719222+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.01361","created_at":"2026-07-05T08:26:38.719222+00:00"},{"alias_kind":"pith_short_12","alias_value":"3CXSQH3WLKBP","created_at":"2026-07-05T08:26:38.719222+00:00"},{"alias_kind":"pith_short_16","alias_value":"3CXSQH3WLKBPMDPN","created_at":"2026-07-05T08:26:38.719222+00:00"},{"alias_kind":"pith_short_8","alias_value":"3CXSQH3W","created_at":"2026-07-05T08:26:38.719222+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.15345","citing_title":"Hadamax Encoding: Elevating Performance in Model-Free Atari","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S","json":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S.json","graph_json":"https://pith.science/api/pith-number/3CXSQH3WLKBPMDPNEWOHSJPR4S/graph.json","events_json":"https://pith.science/api/pith-number/3CXSQH3WLKBPMDPNEWOHSJPR4S/events.json","paper":"https://pith.science/paper/3CXSQH3W"},"agent_actions":{"view_html":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S","download_json":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S.json","view_paper":"https://pith.science/paper/3CXSQH3W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.01361&json=true","fetch_graph":"https://pith.science/api/pith-number/3CXSQH3WLKBPMDPNEWOHSJPR4S/graph.json","fetch_events":"https://pith.science/api/pith-number/3CXSQH3WLKBPMDPNEWOHSJPR4S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S/action/storage_attestation","attest_author":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S/action/author_attestation","sign_citation":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S/action/citation_signature","submit_replication":"https://pith.science/pith/3CXSQH3WLKBPMDPNEWOHSJPR4S/action/replication_record"}},"created_at":"2026-07-05T08:26:38.719222+00:00","updated_at":"2026-07-05T08:26:38.719222+00:00"}