TrajAgent: An LLM-based Agent Framework for Automated Trajectory Modeling via Collaboration of Large and Small Models

2024-10-27

AI Summary

Trajectory modeling faces challenges of high data heterogeneity and task diversity, leading to prohibitive manual modeling costs and poor generalizability. To address this, we propose the first LLM-based agent framework specifically designed for trajectory analysis. Our approach introduces UniEnv—a unified trajectory execution environment—that enables dynamic task decomposition, multi-model collaborative scheduling, and joint optimization. By integrating prompt-driven task orchestration, lightweight model fine-tuning/distillation, and abstracted multi-model interface design, the framework achieves automated pattern discovery and prediction across diverse datasets and tasks. Evaluated on four real-world trajectory datasets across four canonical task categories, our method outperforms state-of-the-art approaches by 2.38%–34.96% in performance. This advancement significantly enhances the universality, scalability, and practical applicability of trajectory modeling.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Modeling other Agents

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Trajectory modeling, which includes research on trajectory data pattern mining and future prediction, has widespread applications in areas such as life services, urban transportation, and public administration. Numerous methods have been proposed to address specific problems within trajectory modeling. However, the heterogeneity of data and the diversity of trajectory tasks make effective and reliable trajectory modeling an important yet highly challenging endeavor, even for domain experts. In this paper, we propose extit{TrajAgent}, a agent framework powered by large language models (LLMs), designed to facilitate robust and efficient trajectory modeling through automation modeling. This framework leverages and optimizes diverse specialized models to address various trajectory modeling tasks across different datasets effectively. In extit{TrajAgent}, we first develop extit{UniEnv}, an execution environment with a unified data and model interface, to support the execution and training of various models. Building on extit{UniEnv}, we introduce an agentic workflow designed for automatic trajectory modeling across various trajectory tasks and data. Furthermore, we introduce collaborative learning schema between LLM-based agents and small speciallized models, to enhance the performance of the whole framework effectively. Extensive experiments on four tasks using four real-world datasets demonstrate the effectiveness of extit{TrajAgent} in automated trajectory modeling, achieving a performance improvement of 2.38%-34.96% over baseline methods.

Problem

• Heterogeneous data and diverse tasks challenge trajectory modeling reliability.
• Automating trajectory modeling across varied datasets and tasks efficiently.
• Enhancing performance via LLM-agent and small model collaboration.

Innovation

• LLM-based agent framework for trajectory modeling
• Unified execution environment for diverse models
• Collaborative learning between LLMs and small models