How Far Are LLMs from Symbolic Planners? An NLP-Based Perspective

TL;DR

This paper evaluates the reasoning capabilities of Large Language Models in AI planning by treating planning as an NLP task, proposing a recovery pipeline, and analyzing its effectiveness compared to classical planners.

Contribution

It introduces an NLP-based evaluation and recovery pipeline for LLM-generated plans, providing a holistic analysis of their planning capabilities and limitations.

Findings

LLMs show no clear evidence of reasoning during plan generation.

The recovery pipeline improves plan quality and success rate.

On average, only 2.65 actions are executable in generated plans.

Abstract

The reasoning and planning abilities of Large Language Models (LLMs) have been a frequent topic of discussion in recent years. Their ability to take unstructured planning problems as input has made LLMs' integration into AI planning an area of interest. Nevertheless, LLMs are still not reliable as planners, with the generated plans often containing mistaken or hallucinated actions. Existing benchmarking and evaluation methods investigate planning with LLMs, focusing primarily on success rate as a quality indicator in various planning tasks, such as validating plans or planning in relaxed conditions. In this paper, we approach planning with LLMs as a natural language processing (NLP) task, given that LLMs are NLP models themselves. We propose a recovery pipeline consisting of an NLP-based evaluation of the generated plans, along with three stages to recover the plans through NLP manipulation of the LLM-generated plans, and eventually complete the plan using a symbolic planner. This pipeline provides a holistic analysis of LLM capabilities in the context of AI task planning, enabling a broader understanding of the quality of invalid plans. Our findings reveal no clear evidence of underlying reasoning during plan generation, and that a pipeline comprising an NLP-based analysis of the plans, followed by a recovery mechanism, still falls short of the quality and reliability of classical planners. On average, only the first 2.65 actions of the plan are executable, with the average length of symbolically generated plans being 8.4 actions. The pipeline still improves action quality and increases the overall success rate from 21.9% to 27.5%.