{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:DO3MJ5XE7VCMC2ZS4Q6AD5WNEU","short_pith_number":"pith:DO3MJ5XE","schema_version":"1.0","canonical_sha256":"1bb6c4f6e4fd44c16b32e43c01f6cd2533109ceab640fbf617228d230b373574","source":{"kind":"arxiv","id":"2301.07608","version":1},"attestation_state":"computed","paper":{"title":"Human-Timescale Adaptation in an Open-Ended Task Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Adaptive Agent Team, Adrian Collister, Alexander Zacherl, Avishkar Bhoopchand, Edward Hughes, Feryal Behbahani, Hannah Openshaw, Jack Parker-Holder, Jakob Bauer, Jakub Sygnowski, Karl Tuyls, Karol Gregor, Kate Baumli, Lei Zhang, Lucy Gonzalez, Maria Loks-Thompson, Michael Chang, Natalie Clay, Nathalie Bradley-Schmieg, Nemanja Rakicevic, Nicolas Perez-Nieves, Sarah York, Satinder Baveja, Sheleem Kashem, Shreya Pathak, Tim Rockt\\\"aschel, Vibhavari Dasagi, Yannick Schroecker","submitted_at":"2023-01-18T15:39:21Z","abstract_excerpt":"Foundation models have shown impressive adaptation and scalability in supervised and self-supervised learning problems, but so far these successes have not fully translated to reinforcement learning (RL). In this work, we demonstrate that training an RL agent at scale leads to a general in-context learning algorithm that can adapt to open-ended novel embodied 3D problems as quickly as humans. In a vast space of held-out environment dynamics, our adaptive agent (AdA) displays on-the-fly hypothesis-driven exploration, efficient exploitation of acquired knowledge, and can successfully be prompted"},"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":"2301.07608","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-01-18T15:39:21Z","cross_cats_sorted":["cs.AI","cs.NE"],"title_canon_sha256":"1c400262df05a15ca799d800ec4d2300d59b6b5311021187055651b3ac0efcae","abstract_canon_sha256":"8a20ad9c8e20aefffe17f4c989e25f6fec23b860726dfad062221d07865faf4c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:34:11.192796Z","signature_b64":"8exbr1smkqqg4g32UTCweHR8YtRvpnWFVOLaoZedjOP1lGufXP6REJJn9qU/rcQIdphCDgFHC6f+ADeqqYwGCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1bb6c4f6e4fd44c16b32e43c01f6cd2533109ceab640fbf617228d230b373574","last_reissued_at":"2026-07-05T05:34:11.192339Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:34:11.192339Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Human-Timescale Adaptation in an Open-Ended Task Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Adaptive Agent Team, Adrian Collister, Alexander Zacherl, Avishkar Bhoopchand, Edward Hughes, Feryal Behbahani, Hannah Openshaw, Jack Parker-Holder, Jakob Bauer, Jakub Sygnowski, Karl Tuyls, Karol Gregor, Kate Baumli, Lei Zhang, Lucy Gonzalez, Maria Loks-Thompson, Michael Chang, Natalie Clay, Nathalie Bradley-Schmieg, Nemanja Rakicevic, Nicolas Perez-Nieves, Sarah York, Satinder Baveja, Sheleem Kashem, Shreya Pathak, Tim Rockt\\\"aschel, Vibhavari Dasagi, Yannick Schroecker","submitted_at":"2023-01-18T15:39:21Z","abstract_excerpt":"Foundation models have shown impressive adaptation and scalability in supervised and self-supervised learning problems, but so far these successes have not fully translated to reinforcement learning (RL). In this work, we demonstrate that training an RL agent at scale leads to a general in-context learning algorithm that can adapt to open-ended novel embodied 3D problems as quickly as humans. In a vast space of held-out environment dynamics, our adaptive agent (AdA) displays on-the-fly hypothesis-driven exploration, efficient exploitation of acquired knowledge, and can successfully be prompted"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.07608","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/2301.07608/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":"2301.07608","created_at":"2026-07-05T05:34:11.192391+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.07608v1","created_at":"2026-07-05T05:34:11.192391+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.07608","created_at":"2026-07-05T05:34:11.192391+00:00"},{"alias_kind":"pith_short_12","alias_value":"DO3MJ5XE7VCM","created_at":"2026-07-05T05:34:11.192391+00:00"},{"alias_kind":"pith_short_16","alias_value":"DO3MJ5XE7VCMC2ZS","created_at":"2026-07-05T05:34:11.192391+00:00"},{"alias_kind":"pith_short_8","alias_value":"DO3MJ5XE","created_at":"2026-07-05T05:34:11.192391+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24962","citing_title":"Towards Scalable Multi-Task Reinforcement Learning with Large Decision Models","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2602.19837","citing_title":"Meta-Learning and Meta-Reinforcement Learning -- Tracing the Path towards DeepMind's Adaptive Agent","ref_index":162,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU","json":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU.json","graph_json":"https://pith.science/api/pith-number/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/graph.json","events_json":"https://pith.science/api/pith-number/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/events.json","paper":"https://pith.science/paper/DO3MJ5XE"},"agent_actions":{"view_html":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU","download_json":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU.json","view_paper":"https://pith.science/paper/DO3MJ5XE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.07608&json=true","fetch_graph":"https://pith.science/api/pith-number/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/graph.json","fetch_events":"https://pith.science/api/pith-number/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/action/storage_attestation","attest_author":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/action/author_attestation","sign_citation":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/action/citation_signature","submit_replication":"https://pith.science/pith/DO3MJ5XE7VCMC2ZS4Q6AD5WNEU/action/replication_record"}},"created_at":"2026-07-05T05:34:11.192391+00:00","updated_at":"2026-07-05T05:34:11.192391+00:00"}