A Survey on Large Language Models for Automated Planning
Mohamed Aghzal, Erion Plaku, Gregory J. Stein, Ziyu Yao
TL;DR
This survey analyzes whether large language models can serve as planning agents, with a focus on long-horizon tasks. It categorizes approaches into standalone planning methods (plan refinement, search-based reasoning, fine-tuning) and integration strategies with traditional planners (text-to-formal translation, commonsense augmentation, plan evaluation). The authors argue that, despite limitations in reliability and efficiency, LLMs offer valuable interfaces and knowledge sources that can enhance planning when combined with symbolic methods, and they identify key challenges—ambiguity, cost, efficiency, knowledge gaps, interpretability, causal reasoning, and multi-agent coordination. The work emphasizes a balanced, hybrid methodology and outlines concrete research directions to advance robust, cost-effective planning systems.
Abstract
The planning ability of Large Language Models (LLMs) has garnered increasing attention in recent years due to their remarkable capacity for multi-step reasoning and their ability to generalize across a wide range of domains. While some researchers emphasize the potential of LLMs to perform complex planning tasks, others highlight significant limitations in their performance, particularly when these models are tasked with handling the intricacies of long-horizon reasoning. In this survey, we critically investigate existing research on the use of LLMs in automated planning, examining both their successes and shortcomings in detail. We illustrate that although LLMs are not well-suited to serve as standalone planners because of these limitations, they nonetheless present an enormous opportunity to enhance planning applications when combined with other approaches. Thus, we advocate for a balanced methodology that leverages the inherent flexibility and generalized knowledge of LLMs alongside the rigor and cost-effectiveness of traditional planning methods.