Embodied Task Planning via Graph-Informed Action Generation with Large Lanaguage Model

2026-01-29

AI Summary

This work addresses the challenges of incoherent planning and hallucination in large language models (LLMs) when deployed for long-horizon task planning in embodied agents, which stem from limited context windows and violations of environmental constraints. To mitigate these issues, the authors propose a Graph-in-Graph memory architecture that encodes environmental states as graph embeddings and organizes them into an execution trajectory graph. This framework integrates structure-aware prior retrieval with a symbolic bounded lookahead module to generate coherent, constraint-compliant plans. Notably, it is the first approach to combine structured graph embedding clustering with symbolic state-transition logic, substantially enhancing LLMs’ long-term planning capabilities in dynamic environments. Evaluated on the Robotouille (synchronous and asynchronous) and ALFWorld benchmarks, the method achieves Pass@1 improvements of 22%, 37%, and 15%, respectively, with comparable or lower computational overhead.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Multiagent Systems: Multiagent Planning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

While Large Language Models (LLMs) have demonstrated strong zero-shot reasoning capabilities, their deployment as embodied agents still faces fundamental challenges in long-horizon planning. Unlike open-ended text generation, embodied agents must decompose high-level intent into actionable sub-goals while strictly adhering to the logic of a dynamic, observed environment. Standard LLM planners frequently fail to maintain strategy coherence over extended horizons due to context window limitation or hallucinate transitions that violate constraints. We propose GiG, a novel planning framework that structures embodied agents'memory using a Graph-in-Graph architecture. Our approach employs a Graph Neural Network (GNN) to encode environmental states into embeddings, organizing these embeddings into action-connected execution trace graphs within an experience memory bank. By clustering these graph embeddings, the framework enables retrieval of structure-aware priors, allowing agents to ground current decisions in relevant past structural patterns. Furthermore, we introduce a novel bounded lookahead module that leverages symbolic transition logic to enhance the agents'planning capabilities through the grounded action projection. We evaluate our framework on three embodied planning benchmarks-Robotouille Synchronous, Robotouille Asynchronous, and ALFWorld. Our method outperforms state-of-the-art baselines, achieving Pass@1 performance gains of up to 22% on Robotouille Synchronous, 37% on Asynchronous, and 15% on ALFWorld with comparable or lower computational cost.

Problem

• Embodied Task Planning
• Long-horizon Planning
• Large Language Models
• Action Generation
• Environment Constraints

Innovation

• Graph-in-Graph architecture
• Embodied task planning
• Graph Neural Network (GNN)
• Structure-aware retrieval
• Bounded lookahead planning