ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2302.01560
30
313

Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents

3 February 2023
Zihao Wang
Shaofei Cai
Guanzhou Chen
Anji Liu
Xiaojian Ma
Yitao Liang
    LM&Ro
    LLMAG
ArXivPDFHTML
Abstract

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 "D‾\underline{D}D​escribe, E‾\underline{E}E​xplain, P‾\underline{P}P​lan and S‾\underline{S}S​elect" (DEPS\textbf{DEPS}DEPS), an interactive planning approach based on Large Language Models (LLMs). DEPS facilitates better error correction on initial LLM-generated plan\textit{plan}plan by integrating description\textit{description}description of the plan execution process and providing self-explanation\textit{explanation}explanation of feedback when encountering failures during the extended planning phases. Furthermore, it includes a goal selector\textit{selector}selector, which is a trainable module that ranks parallel candidate sub-goals based on the estimated steps of completion, consequently refining the initial plan. 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. Further testing reveals our method's general effectiveness in popularly adopted non-open-ended domains as well (i.e., ALFWorld and tabletop manipulation). The ablation and exploratory studies detail how our design beats the counterparts and provide a promising update on the ObtainDiamond\texttt{ObtainDiamond}ObtainDiamond grand challenge with our approach. The code is released at https://github.com/CraftJarvis/MC-Planner.

View on arXiv
Comments on this paper