Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
This addresses the challenge of grounding natural language tasks to executable actions for embodied agents, showing incremental progress by enhancing plan executability.
The paper tackles the problem of using large language models (LLMs) to decompose high-level tasks into actionable steps for embodied agents, finding that with appropriate prompting, LLMs can generate mid-level plans without training, and a proposed method improves executability over baselines in the VirtualHome environment.
Can world knowledge learned by large language models (LLMs) be used to act in interactive environments? In this paper, we investigate the possibility of grounding high-level tasks, expressed in natural language (e.g. "make breakfast"), to a chosen set of actionable steps (e.g. "open fridge"). While prior work focused on learning from explicit step-by-step examples of how to act, we surprisingly find that if pre-trained LMs are large enough and prompted appropriately, they can effectively decompose high-level tasks into mid-level plans without any further training. However, the plans produced naively by LLMs often cannot map precisely to admissible actions. We propose a procedure that conditions on existing demonstrations and semantically translates the plans to admissible actions. Our evaluation in the recent VirtualHome environment shows that the resulting method substantially improves executability over the LLM baseline. The conducted human evaluation reveals a trade-off between executability and correctness but shows a promising sign towards extracting actionable knowledge from language models. Website at https://huangwl18.github.io/language-planner