{"paper":{"title":"Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"DEPS combines LLM-based interactive planning with a trainable goal selector to create a zero-shot multi-task agent that completes 70+ Minecraft tasks and nearly doubles prior performance.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Anji Liu, Guanzhou Chen, Shaofei Cai, Xiaojian Ma, Yitao Liang, Zihao Wang","submitted_at":"2023-02-03T06:06:27Z","abstract_excerpt":"We investigate the challenge of task planning for multi-task embodied agents in open-world environments. Two main difficulties are identified: 1) executing plans in an open-world environment (e.g., Minecraft) necessitates accurate and multi-step reasoning due to the long-term nature of tasks, and 2) as vanilla planners do not consider how easy the current agent can achieve a given sub-task when ordering parallel sub-goals within a complicated plan, the resulting plan could be inefficient or even infeasible. To this end, we propose \"$\\underline{D}$escribe, $\\underline{E}$xplain, $\\underline{P}$"},"claims":{"count":3,"items":[{"kind":"strongest_claim","text":"Our experiments mark the milestone of the first zero-shot multi-task agent that can robustly accomplish 70+ Minecraft tasks and nearly double the overall performances.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That large language models can reliably produce accurate descriptions and self-explanations of plan failures without hallucination, and that the trained goal selector generalizes to estimate completion steps across diverse tasks.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DEPS combines LLM-based interactive planning with a trainable goal selector to create a zero-shot multi-task agent that completes 70+ Minecraft tasks and nearly doubles prior performance.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"}],"snapshot_sha256":"5468a78e3ef7478e80f22524291f86337e301b2d129c82ec0b06c84469ad86d1"},"source":{"id":"2302.01560","kind":"arxiv","version":3},"verdict":{"id":"cd7dafa9-292a-4bbe-82b3-f81a11658453","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T03:20:17.333742Z","strongest_claim":"Our experiments mark the milestone of the first zero-shot multi-task agent that can robustly accomplish 70+ Minecraft tasks and nearly double the overall performances.","one_line_summary":"DEPS combines LLM-based interactive planning with a trainable goal selector to create a zero-shot multi-task agent that completes 70+ Minecraft tasks and nearly doubles prior performance.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That large language models can reliably produce accurate descriptions and self-explanations of plan failures without hallucination, and that the trained goal selector generalizes to estimate completion steps across diverse tasks.","pith_extraction_headline":""},"references":{"count":71,"sample":[{"doi":"","year":2022,"title":"Flamingo: a Visual Language Model for Few-Shot Learning","work_id":"a110f764-38dc-41b2-a802-53744ecea1fc","ref_index":1,"cited_arxiv_id":"2204.14198","is_internal_anchor":true},{"doi":"","year":2017,"title":"P.-L. Bacon, J. Harb, and D. Precup. The option-critic architecture. In Proceedings of the AAAI conference on artificial intelligence, 2017. 1","work_id":"95e5f834-6420-4a4e-a0a0-c797157ae93e","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Video pretraining (vpt): Learning to act by watching unlabeled online videos","work_id":"2bf44de9-4b8f-4ab3-816e-2ffb5278c578","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"A. Brohan, Y . Chebotar, C. Finn, K. Hausman, A. Herzog, D. Ho, J. Ibarz, A. Irpan, E. Jang, R. Julian, et al. Do as i can, not as i say: Grounding language in robotic affordances. In 6th Annual Confe","work_id":"970cbc3b-0b27-4b39-b3fc-d86e1bf02280","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1901,"title":"T. Brown, B. Mann, N. Ryder, M. Subbiah, J. D. Kaplan, P. Dhariwal, A. Neelakantan, P. Shyam, G. Sastry, A. Askell, et al. Language models are few-shot learners. Advances in neural information process","work_id":"bc17df25-2629-4184-aac1-f5bb84bd1856","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":71,"snapshot_sha256":"e9dac2d608cc2a91f80f55beaeb0d8f60e6ed49f19f98ad20b562ce4f2a4c63f","internal_anchors":18},"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"}