CoMaPOI: A Collaborative Multi-Agent Framework for Next POI Prediction Bridging the Gap Between Trajectory and Language

2025-05-28

AI Summary

Large language models (LLMs) face two key bottlenecks in next point-of-interest (POI) prediction: (1) limited understanding of numerical spatiotemporal trajectories, and (2) prediction bias arising from excessively large candidate spaces. To address these, we propose the first multi-agent collaborative framework tailored for spatiotemporal tasks—comprising a Profiler that semantically encodes raw trajectories, a Forecaster that dynamically prunes the candidate set based on evolving user preferences, and a Predictor that jointly leverages spatiotemporal semantics and LLM-based linguistic reasoning for accurate prediction. Our framework achieves deep coupling between trajectory modeling and LLM inference, integrating spatiotemporal feature engineering, semantic transformation, dynamic candidate constraint, and prompt-driven joint reasoning. Evaluated on three standard benchmarks—NYC, TKY, and CA—it consistently outperforms state-of-the-art methods, with improvements of 5–10% across all metrics, demonstrating the effectiveness and generalizability of multi-agent collaboration in spatiotemporal–language joint modeling.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

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

Abstract

Large Language Models (LLMs) offer new opportunities for the next Point-Of-Interest (POI) prediction task, leveraging their capabilities in semantic understanding of POI trajectories. However, previous LLM-based methods, which are superficially adapted to next POI prediction, largely overlook critical challenges associated with applying LLMs to this task. Specifically, LLMs encounter two critical challenges: (1) a lack of intrinsic understanding of numeric spatiotemporal data, which hinders accurate modeling of users' spatiotemporal distributions and preferences; and (2) an excessively large and unconstrained candidate POI space, which often results in random or irrelevant predictions. To address these issues, we propose a Collaborative Multi Agent Framework for Next POI Prediction, named CoMaPOI. Through the close interaction of three specialized agents (Profiler, Forecaster, and Predictor), CoMaPOI collaboratively addresses the two critical challenges. The Profiler agent is responsible for converting numeric data into language descriptions, enhancing semantic understanding. The Forecaster agent focuses on dynamically constraining and refining the candidate POI space. The Predictor agent integrates this information to generate high-precision predictions. Extensive experiments on three benchmark datasets (NYC, TKY, and CA) demonstrate that CoMaPOI achieves state of the art performance, improving all metrics by 5% to 10% compared to SOTA baselines. This work pioneers the investigation of challenges associated with applying LLMs to complex spatiotemporal tasks by leveraging tailored collaborative agents.

Problem

• LLMs lack understanding of numeric spatiotemporal data in POI prediction
• Unconstrained candidate POI space leads to irrelevant predictions
• Proposes multi-agent framework to enhance POI prediction accuracy

Innovation

• Collaborative multi-agent framework for POI prediction
• Converts numeric data into language descriptions
• Dynamically constrains candidate POI space