{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ROXZ4YP4EYRTTZTPFV5DEB27NH","short_pith_number":"pith:ROXZ4YP4","schema_version":"1.0","canonical_sha256":"8baf9e61fc262339e66f2d7a32075f69f432dcbc86b327c4df166d0d111935e2","source":{"kind":"arxiv","id":"2408.15950","version":2},"attestation_state":"computed","paper":{"title":"Atari-GPT: Benchmarking Multimodal Large Language Models as Low-Level Policies in Atari Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Devin White, MD Sunbeam, Nicholas R. Waytowich, Vinicius G. Goecks","submitted_at":"2024-08-28T17:08:56Z","abstract_excerpt":"Recent advancements in large language models (LLMs) have expanded their capabilities beyond traditional text-based tasks to multimodal domains, integrating visual, auditory, and textual data. While multimodal LLMs have been extensively explored for high-level planning in domains like robotics and games, their potential as low-level controllers remains largely untapped. In this paper, we introduce a novel benchmark aimed at testing the emergent capabilities of multimodal LLMs as low-level policies in Atari games. Unlike traditional reinforcement learning (RL) methods that require training for e"},"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":"2408.15950","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-08-28T17:08:56Z","cross_cats_sorted":[],"title_canon_sha256":"e3fd94fef78c772f56d91e0552255c1406c8fe1d7fc3d8b004a6a6961c9e3d34","abstract_canon_sha256":"c54a2d33e02b1918d35b7f2ad8da2cdf48d52e5d4a084dc7a9d4994b5063bdb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:42:29.638124Z","signature_b64":"md3cqC3Jh+4kyJSr4+kWTXJyDvjDQA/PbNgcSgBJZtxcBNTnt1csaL1G+DYl2FGQ80ffTW0kngm5YJdb1XGrCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8baf9e61fc262339e66f2d7a32075f69f432dcbc86b327c4df166d0d111935e2","last_reissued_at":"2026-07-05T09:42:29.637601Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:42:29.637601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Atari-GPT: Benchmarking Multimodal Large Language Models as Low-Level Policies in Atari Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Devin White, MD Sunbeam, Nicholas R. Waytowich, Vinicius G. Goecks","submitted_at":"2024-08-28T17:08:56Z","abstract_excerpt":"Recent advancements in large language models (LLMs) have expanded their capabilities beyond traditional text-based tasks to multimodal domains, integrating visual, auditory, and textual data. While multimodal LLMs have been extensively explored for high-level planning in domains like robotics and games, their potential as low-level controllers remains largely untapped. In this paper, we introduce a novel benchmark aimed at testing the emergent capabilities of multimodal LLMs as low-level policies in Atari games. Unlike traditional reinforcement learning (RL) methods that require training for e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.15950","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/2408.15950/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":"2408.15950","created_at":"2026-07-05T09:42:29.637673+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.15950v2","created_at":"2026-07-05T09:42:29.637673+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.15950","created_at":"2026-07-05T09:42:29.637673+00:00"},{"alias_kind":"pith_short_12","alias_value":"ROXZ4YP4EYRT","created_at":"2026-07-05T09:42:29.637673+00:00"},{"alias_kind":"pith_short_16","alias_value":"ROXZ4YP4EYRTTZTP","created_at":"2026-07-05T09:42:29.637673+00:00"},{"alias_kind":"pith_short_8","alias_value":"ROXZ4YP4","created_at":"2026-07-05T09:42:29.637673+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08233","citing_title":"Playing ZendoWorld: Challenging AI Agents on Active Visual Concept Induction","ref_index":34,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH","json":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH.json","graph_json":"https://pith.science/api/pith-number/ROXZ4YP4EYRTTZTPFV5DEB27NH/graph.json","events_json":"https://pith.science/api/pith-number/ROXZ4YP4EYRTTZTPFV5DEB27NH/events.json","paper":"https://pith.science/paper/ROXZ4YP4"},"agent_actions":{"view_html":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH","download_json":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH.json","view_paper":"https://pith.science/paper/ROXZ4YP4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.15950&json=true","fetch_graph":"https://pith.science/api/pith-number/ROXZ4YP4EYRTTZTPFV5DEB27NH/graph.json","fetch_events":"https://pith.science/api/pith-number/ROXZ4YP4EYRTTZTPFV5DEB27NH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH/action/storage_attestation","attest_author":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH/action/author_attestation","sign_citation":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH/action/citation_signature","submit_replication":"https://pith.science/pith/ROXZ4YP4EYRTTZTPFV5DEB27NH/action/replication_record"}},"created_at":"2026-07-05T09:42:29.637673+00:00","updated_at":"2026-07-05T09:42:29.637673+00:00"}