Dynamic Planning with a LLM

TL;DR

This paper introduces LLM-DP, a neuro-symbolic framework combining LLMs with traditional planners to improve multi-step reasoning and efficiency in embodied task planning.

Contribution

It presents a novel neuro-symbolic framework that integrates LLMs with symbolic planning to enhance efficiency and accuracy in complex embodied tasks.

Findings

LLM-DP outperforms naive LLM ReAct baseline in Alfworld.

LLM-DP is faster and more efficient in solving embodied planning tasks.

The framework effectively handles noisy observations and uncertainty.

Abstract

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.