The paper presents a comprehensive review of 126 papers, investigating eight categories based on the unique applications of Large Language Models (LLMs) in addressing various aspects of planning problems. The eight categories include language translation, plan generation, model construction, multi-agent planning, interactive planning, heuristics optimization, tool integration, and brain-inspired planning. The study reveals that the true potential of LLMs unfolds when they are integrated with traditional symbolic planners, offering a promising neuro-symbolic approach that combines the generative aspects of LLMs with the precision of classical planning methods. The integration of LLMs in Automated Planning and Scheduling (APS) could address complex planning challenges, advocating for a direction in automated planning that leverages these synergistic capabilities to develop more advanced and intelligent planning systems.
Publication date: 8 Jan 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2401.02500
