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

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.

References in corpus (95)

• Tree of Thoughts: Deliberate Problem Solving with Large Language Models
• Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
• Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
• Graph of Thoughts: Solving Elaborate Problems with Large Language Models
• PaLM-E: An Embodied Multimodal Language Model
• HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
• Text2Motion: From Natural Language Instructions to Feasible Plans
• Inner Monologue: Embodied Reasoning through Planning with Language Models
• Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents
• Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models
• VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models
• LLM+P: Empowering Large Language Models with Optimal Planning Proficiency
• Large Language Models as Zero-Shot Human Models for Human-Robot Interaction
• OpenAGI: When LLM Meets Domain Experts
• Cognitive Architectures for Language Agents
• Integrating Action Knowledge and LLMs for Task Planning and Situation Handling in Open Worlds
• Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents
• PlanBench: An Extensible Benchmark for Evaluating Large Language Models on Planning and Reasoning about Change
• Translating Natural Language to Planning Goals with Large-Language Models
• Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners
• Large Language Models as Commonsense Knowledge for Large-Scale Task Planning
• Building Cooperative Embodied Agents Modularly with Large Language Models
• Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning
• From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thought
• Leveraging Pre-trained Large Language Models to Construct and Utilize World Models for Model-based Task Planning
• On the Planning Abilities of Large Language Models (A Critical Investigation with a Proposed Benchmark)
• CoPAL: Corrective Planning of Robot Actions with Large Language Models
• SayPlan: Grounding Large Language Models using 3D Scene Graphs for Scalable Robot Task Planning
• SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks
• Getting from Generative AI to Trustworthy AI: What LLMs might learn from Cyc
• Evaluating Cognitive Maps and Planning in Large Language Models with CogEval
• ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings
• REFLECT: Summarizing Robot Experiences for Failure Explanation and Correction
• Embodied Task Planning with Large Language Models
• Large Language Models are In-Context Semantic Reasoners rather than Symbolic Reasoners
• Parsel: Algorithmic Reasoning with Language Models by Composing Decompositions
• Navigation with Large Language Models: Semantic Guesswork as a Heuristic for Planning
• Grounded Decoding: Guiding Text Generation with Grounded Models for Embodied Agents
• War and Peace (WarAgent): Large Language Model-based Multi-Agent Simulation of World Wars
• API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs
• AdaPlanner: Adaptive Planning from Feedback with Language Models
• TPTU: Large Language Model-based AI Agents for Task Planning and Tool Usage
• Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive Negotiation
• Enabling Intelligent Interactions between an Agent and an LLM: A Reinforcement Learning Approach
• ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planning
• Plansformer: Generating Symbolic Plans using Transformers
• SMART-LLM: Smart Multi-Agent Robot Task Planning using Large Language Models
• Skill Reinforcement Learning and Planning for Open-World Long-Horizon Tasks
• Can Large Language Models Really Improve by Self-critiquing Their Own Plans?
• AutoTAMP: Autoregressive Task and Motion Planning with LLMs as Translators and Checkers
• VideoDirectorGPT: Consistent Multi-scene Video Generation via LLM-Guided Planning
• Tool Documentation Enables Zero-Shot Tool-Usage with Large Language Models
• Do Embodied Agents Dream of Pixelated Sheep: Embodied Decision Making using Language Guided World Modelling
• Halo: Estimation and Reduction of Hallucinations in Open-Source Weak Large Language Models
• Statler: State-Maintaining Language Models for Embodied Reasoning
• Strategic Reasoning with Language Models
• Gentopia: A Collaborative Platform for Tool-Augmented LLMs
• Chain-of-Symbol Prompting Elicits Planning in Large Langauge Models
• Scalable Multi-Robot Collaboration with Large Language Models: Centralized or Decentralized Systems?
• Bootstrap Your Own Skills: Learning to Solve New Tasks with Large Language Model Guidance
• Plan, Eliminate, and Track -- Language Models are Good Teachers for Embodied Agents
• Understanding the Capabilities of Large Language Models for Automated Planning
• Tree-Planner: Efficient Close-loop Task Planning with Large Language Models
• PREFER: Prompt Ensemble Learning via Feedback-Reflect-Refine
• Plan-Seq-Learn: Language Model Guided RL for Solving Long Horizon Robotics Tasks
• OceanChat: Piloting Autonomous Underwater Vehicles in Natural Language
• Conformal Temporal Logic Planning using Large Language Models
• Tree-of-Mixed-Thought: Combining Fast and Slow Thinking for Multi-hop Visual Reasoning
• Optimal Scene Graph Planning with Large Language Model Guidance
• DynaCon: Dynamic Robot Planner with Contextual Awareness via LLMs
• Alphazero-like Tree-Search can Guide Large Language Model Decoding and Training
• Diversity of Thought Improves Reasoning Abilities of LLMs
• Learning and Leveraging Verifiers to Improve Planning Capabilities of Pre-trained Language Models
• Steve-Eye: Equipping LLM-based Embodied Agents with Visual Perception in Open Worlds
• From Static to Dynamic: A Continual Learning Framework for Large Language Models
• Vision-Language Interpreter for Robot Task Planning
• Multimodal Procedural Planning via Dual Text-Image Prompting
• Automaton-Based Representations of Task Knowledge from Generative Language Models
• Neuro Symbolic Reasoning for Planning: Counterexample Guided Inductive Synthesis using Large Language Models and Satisfiability Solving
• ISR-LLM: Iterative Self-Refined Large Language Model for Long-Horizon Sequential Task Planning
• Asking Before Acting: Gather Information in Embodied Decision Making with Language Models
• Creative Robot Tool Use with Large Language Models
• A Versatile Graph Learning Approach through LLM-based Agent
• Improving Planning with Large Language Models: A Modular Agentic Architecture
• Planning with Logical Graph-based Language Model for Instruction Generation
• Fast and Slow Planning
• Lifelong Robot Learning with Human Assisted Language Planners
• Guiding Language Model Reasoning with Planning Tokens
• EIPE-text: Evaluation-Guided Iterative Plan Extraction for Long-Form Narrative Text Generation
• Language Models, Agent Models, and World Models: The LAW for Machine Reasoning and Planning
• LLM-Grounder: Open-Vocabulary 3D Visual Grounding with Large Language Model as an Agent
• On the Planning, Search, and Memorization Capabilities of Large Language Models
• Human-Centered Planning
• Dynamic Planning with a LLM
• Improving Generalization in Task-oriented Dialogues with Workflows and Action Plans

Cited by in corpus (3)

• An investigation on the use of Large Language Models for hyperparameter tuning in Evolutionary Algorithms
• VeriPlan: Integrating Formal Verification and LLMs into End-User Planning
• FISHNET: Financial Intelligence from Sub-querying, Harmonizing, Neural-Conditioning, Expert Swarms, and Task Planning