Where You Go is Who You Are: Behavioral Theory-Guided LLMs for Inverse Reinforcement Learning

2025-05-22

AI Summary

To address low accuracy and insufficient cognitive modeling in inferring sociodemographic attributes from trajectory data, this paper proposes a novel framework integrating the Theory of Planned Behavior (TPB) with Inverse Reinforcement Learning (IRL). It is the first to deeply embed TPB into IRL by designing a cognition-driven reward function grounded in behavioral decision-making principles. We further introduce a Large Language Model (LLM)-guided Chain-of-Cognitive-Reasoning (CCR) mechanism to enable interpretable mapping from travel intentions to sociodemographic attributes. Additionally, LLMs are leveraged to optimize reward initialization and iterative updates, mitigating IRL’s ill-posedness. Experiments on the Puget Sound Transportation Survey dataset demonstrate that our method improves sociodemographic attribute prediction accuracy by 18.7% over state-of-the-art approaches, significantly enhancing the behavioral interpretability of trajectory data and its utility for evidence-based transportation policy design.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Knowledge Representation and Reasoning: Preferences

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: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Big trajectory data hold great promise for human mobility analysis, but their utility is often constrained by the absence of critical traveler attributes, particularly sociodemographic information. While prior studies have explored predicting such attributes from mobility patterns, they often overlooked underlying cognitive mechanisms and exhibited low predictive accuracy. This study introduces SILIC, short for Sociodemographic Inference with LLM-guided Inverse Reinforcement Learning (IRL) and Cognitive Chain Reasoning (CCR), a theoretically grounded framework that leverages LLMs to infer sociodemographic attributes from observed mobility patterns by capturing latent behavioral intentions and reasoning through psychological constructs. Particularly, our approach explicitly follows the Theory of Planned Behavior (TPB), a foundational behavioral framework in transportation research, to model individuals' latent cognitive processes underlying travel decision-making. The LLMs further provide heuristic guidance to improve IRL reward function initialization and update by addressing its ill-posedness and optimization challenges arising from the vast and unstructured reward space. Evaluated in the 2017 Puget Sound Regional Council Household Travel Survey, our method substantially outperforms state-of-the-art baselines and shows great promise for enriching big trajectory data to support more behaviorally grounded applications in transportation planning and beyond.

Problem

• Inferring sociodemographic attributes from mobility patterns
• Modeling latent cognitive processes in travel decisions
• Improving inverse reinforcement learning reward functions

Innovation

• LLM-guided Inverse Reinforcement Learning for sociodemographic inference
• Cognitive Chain Reasoning to model behavioral intentions
• Theory of Planned Behavior for travel decision-making