Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

On the Prospects of Incorporating Large Language Models (LLMs) in Automated Planning and Scheduling (APS)

Vishal Pallagani, Kaushik Roy, Bharath Muppasani, Francesco Fabiano, Andrea Loreggia, Keerthiram Murugesan, Biplav Srivastava, Francesca Rossi, Lior Horesh, Amit Sheth

TL;DR

The paper surveys the integration of Large Language Models (LLMs) into Automated Planning and Scheduling (APS) across eight categories, based on 126 recent works. It argues for a neuro-symbolic future where LLMs complement symbolic planners, combining language-enabled reasoning with precise plan generation. Key findings highlight both the strengths of LLMs in language interfaces and knowledge grounding, and the limitations in producing robust, scalable plans without symbolic back-ends. The work outlines future research directions, including new training paradigms, explicit neuro-symbolic taxonomies, and standardized metrics to evaluate LLM-assisted planners, with practical implications for the ICAPS community.

Abstract

Automated Planning and Scheduling is among the growing areas in Artificial Intelligence (AI) where mention of LLMs has gained popularity. Based on a comprehensive review of 126 papers, this paper investigates eight categories based on the unique applications of LLMs in addressing various aspects of planning problems: language translation, plan generation, model construction, multi-agent planning, interactive planning, heuristics optimization, tool integration, and brain-inspired planning. For each category, we articulate the issues considered and existing gaps. A critical insight resulting from our review is that the true potential of LLMs unfolds when they are integrated with traditional symbolic planners, pointing towards a promising neuro-symbolic approach. This approach effectively combines the generative aspects of LLMs with the precision of classical planning methods. By synthesizing insights from existing literature, we underline the potential of this integration to address complex planning challenges. Our goal is to encourage the ICAPS community to recognize the complementary strengths of LLMs and symbolic planners, advocating for a direction in automated planning that leverages these synergistic capabilities to develop more advanced and intelligent planning systems.

On the Prospects of Incorporating Large Language Models (LLMs) in Automated Planning and Scheduling (APS)

TL;DR

The paper surveys the integration of Large Language Models (LLMs) into Automated Planning and Scheduling (APS) across eight categories, based on 126 recent works. It argues for a neuro-symbolic future where LLMs complement symbolic planners, combining language-enabled reasoning with precise plan generation. Key findings highlight both the strengths of LLMs in language interfaces and knowledge grounding, and the limitations in producing robust, scalable plans without symbolic back-ends. The work outlines future research directions, including new training paradigms, explicit neuro-symbolic taxonomies, and standardized metrics to evaluate LLM-assisted planners, with practical implications for the ICAPS community.

Abstract

Automated Planning and Scheduling is among the growing areas in Artificial Intelligence (AI) where mention of LLMs has gained popularity. Based on a comprehensive review of 126 papers, this paper investigates eight categories based on the unique applications of LLMs in addressing various aspects of planning problems: language translation, plan generation, model construction, multi-agent planning, interactive planning, heuristics optimization, tool integration, and brain-inspired planning. For each category, we articulate the issues considered and existing gaps. A critical insight resulting from our review is that the true potential of LLMs unfolds when they are integrated with traditional symbolic planners, pointing towards a promising neuro-symbolic approach. This approach effectively combines the generative aspects of LLMs with the precision of classical planning methods. By synthesizing insights from existing literature, we underline the potential of this integration to address complex planning challenges. Our goal is to encourage the ICAPS community to recognize the complementary strengths of LLMs and symbolic planners, advocating for a direction in automated planning that leverages these synergistic capabilities to develop more advanced and intelligent planning systems.
Paper Structure (15 sections, 6 equations, 5 figures, 1 table)

This paper contains 15 sections, 6 equations, 5 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. Large Language Models
  4. Automated Planning and Scheduling
  5. LLMs in APS -- Literature selection
  6. LLMs in APS -- Literature Discussion
  7. Language Translation
  8. Plan Generation
  9. Model Construction
  10. Multi-agent Planning
  11. Interactive Planning
  12. Heuristics Optimization
  13. Tool Integration
  14. Brain-Inspired Planning
  15. Discussion and Conclusion

Figures (5)

  • Figure 1: Radar chart showcasing the relative performance of six language models (GPT-4, Claude-v1, GPT-3.5-turbo, Vicuna-13B, Alpaca-13B, LLama-13B) across key domains: Writing, Roleplay, Reasoning, Math, Coding, Extraction, STEM, and Humanities from zheng2023judging.
  • Figure 2: Of the 126 papers surveyed in this study, 55 were accepted by peer-reviewed conferences. This chart illustrates the distribution of these papers across various conferences in the fields of LLMs and APS, highlighting the primary forums for scholarly contributions in these areas.
  • Figure 3: Annual distribution of the 126 surveyed papers, indicating a significant increase in publications from 12 in 2022 to 114 in 2023, highlighting the rapid growth of LLM research within a single year.
  • Figure 4: Word cloud of terms from the titles of papers surveyed in this study, displaying the prevalence of "Language Model" and "Planning" as central themes. The presence of "Neuro-Symbolic" indicates an emergent trend toward the fusion of neural and symbolic methodologies in the domain.
  • Figure 5: Taxonomy of recent research in the intersection of LLMs and Planning into categories (#). Each has scholarly papers based on their unique application or customization of LLMs in addressing various aspects of planning problems.