{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:HSTQG4I3XOTAOVH7WZHEERMJTX","short_pith_number":"pith:HSTQG4I3","schema_version":"1.0","canonical_sha256":"3ca703711bbba60754ffb64e4245899de4879c4725056c4be0a25f2bfa8ec43c","source":{"kind":"arxiv","id":"2301.12050","version":2},"attestation_state":"computed","paper":{"title":"Do Embodied Agents Dream of Pixelated Sheep: Embodied Decision Making using Language Guided World Modelling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Alane Suhr, Hannaneh Hajishirzi, Kolby Nottingham, Prithviraj Ammanabrolu, Roy Fox, Sameer Singh, Yejin Choi","submitted_at":"2023-01-28T02:04:07Z","abstract_excerpt":"Reinforcement learning (RL) agents typically learn tabula rasa, without prior knowledge of the world. However, if initialized with knowledge of high-level subgoals and transitions between subgoals, RL agents could utilize this Abstract World Model (AWM) for planning and exploration. We propose using few-shot large language models (LLMs) to hypothesize an AWM, that will be verified through world experience, to improve sample efficiency of RL agents. Our DECKARD agent applies LLM-guided exploration to item crafting in Minecraft in two phases: (1) the Dream phase where the agent uses an LLM to de"},"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":"2301.12050","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-28T02:04:07Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"027b2b3c33233c5c777b54c82f580629cf89ccb746fca65371b2d984d11f6e18","abstract_canon_sha256":"df505377e413d4367e824f5211e89e03127babcdad70fa6216a79e4dfa4f7850"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:04:53.960386Z","signature_b64":"MJ6YmHR0s9iUUvc0O0S8H7qrNXouvoxxJIru8567GgVBroLZAfkhm3/CqsLd0ficlE+7hlZz6cFPdzlEBY0RBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ca703711bbba60754ffb64e4245899de4879c4725056c4be0a25f2bfa8ec43c","last_reissued_at":"2026-07-05T06:04:53.959861Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:04:53.959861Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Embodied Agents Dream of Pixelated Sheep: Embodied Decision Making using Language Guided World Modelling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Alane Suhr, Hannaneh Hajishirzi, Kolby Nottingham, Prithviraj Ammanabrolu, Roy Fox, Sameer Singh, Yejin Choi","submitted_at":"2023-01-28T02:04:07Z","abstract_excerpt":"Reinforcement learning (RL) agents typically learn tabula rasa, without prior knowledge of the world. However, if initialized with knowledge of high-level subgoals and transitions between subgoals, RL agents could utilize this Abstract World Model (AWM) for planning and exploration. We propose using few-shot large language models (LLMs) to hypothesize an AWM, that will be verified through world experience, to improve sample efficiency of RL agents. Our DECKARD agent applies LLM-guided exploration to item crafting in Minecraft in two phases: (1) the Dream phase where the agent uses an LLM to de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.12050","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/2301.12050/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":"2301.12050","created_at":"2026-07-05T06:04:53.959918+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.12050v2","created_at":"2026-07-05T06:04:53.959918+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.12050","created_at":"2026-07-05T06:04:53.959918+00:00"},{"alias_kind":"pith_short_12","alias_value":"HSTQG4I3XOTA","created_at":"2026-07-05T06:04:53.959918+00:00"},{"alias_kind":"pith_short_16","alias_value":"HSTQG4I3XOTAOVH7","created_at":"2026-07-05T06:04:53.959918+00:00"},{"alias_kind":"pith_short_8","alias_value":"HSTQG4I3","created_at":"2026-07-05T06:04:53.959918+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27806","citing_title":"Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27806","citing_title":"Agent vs. Parametric World Models: Hybrid Planning for Reliable Language Agents","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2303.17491","citing_title":"Language Models can Solve Computer Tasks","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2602.20867","citing_title":"SoK: Agentic Skills -- Beyond Tool Use in LLM Agents","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2303.03378","citing_title":"PaLM-E: An Embodied Multimodal Language Model","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX","json":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX.json","graph_json":"https://pith.science/api/pith-number/HSTQG4I3XOTAOVH7WZHEERMJTX/graph.json","events_json":"https://pith.science/api/pith-number/HSTQG4I3XOTAOVH7WZHEERMJTX/events.json","paper":"https://pith.science/paper/HSTQG4I3"},"agent_actions":{"view_html":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX","download_json":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX.json","view_paper":"https://pith.science/paper/HSTQG4I3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.12050&json=true","fetch_graph":"https://pith.science/api/pith-number/HSTQG4I3XOTAOVH7WZHEERMJTX/graph.json","fetch_events":"https://pith.science/api/pith-number/HSTQG4I3XOTAOVH7WZHEERMJTX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX/action/storage_attestation","attest_author":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX/action/author_attestation","sign_citation":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX/action/citation_signature","submit_replication":"https://pith.science/pith/HSTQG4I3XOTAOVH7WZHEERMJTX/action/replication_record"}},"created_at":"2026-07-05T06:04:53.959918+00:00","updated_at":"2026-07-05T06:04:53.959918+00:00"}