Planning in the LLM Era: Building for Reliability and Efficiency

TL;DR

This paper reviews recent shifts in planning methods using large language models, emphasizing reliability and efficiency in agent design by generating symbolic planners for verification and resource-efficient inference.

Contribution

It analyzes three major categories of planner-generation methods, discusses their limitations, and proposes research directions for more reliable and efficient LLM-based planning.

Findings

Early LLM-based planning relied on single-shot generation and hybrid search methods.

Recent approaches generate symbolic solvers verified for efficient inference.

Shift towards LLM-generated symbolic planners enhances reliability and resource efficiency.

Abstract

Growing attention to intelligent agents has put a spotlight on one of their central capabilities: planning. Early attempts to leverage large language models (LLMs) for planning relied on single-shot plan generation, followed by hybrid approaches that coupled LLMs with limited external search. These methods, unsound and incomplete by their very nature, often require substantial resources without yielding better solutions on unseen problems. As the limitations of LLMs become clearer, recent work has shifted toward using them at solution construction time -- generating symbolic solvers for a family of problems that can be verified and then used efficiently at inference time. This trend reflects the growing need for agents that are both reliable and resource-efficient. It also offers a path towards generating maintainable planners with minimal dependence on language models at inference time. In this paper, we argue that this shift reflects a broader realignment of the planning field in the LLM era. We examine three major categories of planner-generation methods, discuss their current limitations, and outline research steps towards a more reliable and efficient LLM-based generation of planners.