Dynamic Planning with a LLM
This addresses the challenge of planning in noisy, uncertain environments for embodied AI applications, though it is incremental as it builds on existing neuro-symbolic approaches.
The paper tackles the problem of enabling embodied agents to perform complex multi-step reasoning by combining an LLM with a traditional planner, resulting in LLM-DP solving Alfworld faster and more efficiently than a naive LLM ReAct baseline.
While Large Language Models (LLMs) can solve many NLP tasks in zero-shot settings, applications involving embodied agents remain problematic. In particular, complex plans that require multi-step reasoning become difficult and too costly as the context window grows. Planning requires understanding the likely effects of one's actions and identifying whether the current environment satisfies the goal state. While symbolic planners find optimal solutions quickly, they require a complete and accurate representation of the planning problem, severely limiting their use in practical scenarios. In contrast, modern LLMs cope with noisy observations and high levels of uncertainty when reasoning about a task. Our work presents LLM Dynamic Planner (LLM-DP): a neuro-symbolic framework where an LLM works hand-in-hand with a traditional planner to solve an embodied task. Given action-descriptions, LLM-DP solves Alfworld faster and more efficiently than a naive LLM ReAct baseline.