arXiv:2511.00993v1 Announce Type: cross 
Abstract: Effective modeling of how human travelers learn and adjust their travel behavior from interacting with transportation systems is critical for system assessment and planning. However, this task is also difficult due to the complex cognition and decision-making involved in such behavior. Recent research has begun to leverage Large Language Model (LLM) agents for this task. Building on this, we introduce a novel dual-agent framework that enables continuous learning and alignment between LLM agents and human travelers on learning and adaptation behavior from online data streams. Our approach involves a set of LLM traveler agents, equipped with a memory system and a learnable persona, which serve as simulators for human travelers. To ensure behavioral alignment, we introduce an LLM calibration agent that leverages the reasoning and analytical capabilities of LLMs to train the personas of these traveler agents. Working together, this dual-agent system is designed to track and align the underlying decision-making mechanisms of travelers and produce realistic, adaptive simulations. Using a real-world dataset from a day-to-day route choice experiment, we show our approach significantly outperforms existing LLM-based methods in both individual behavioral alignment and aggregate simulation accuracy. Furthermore, we demonstrate that our method moves beyond simple behavioral mimicry to capture the evolution of underlying learning processes, a deeper alignment that fosters robust generalization. Overall, our framework provides a new approach for creating adaptive and behaviorally realistic agents to simulate travelers' learning and adaptation that can benefit transportation simulation and policy analysis.

تقدم دراسة حديثة إطار عمل مزدوج الوكيل يعزز كيفية مساعدة وكلاء نماذج اللغة الكبيرة (LLM) في فهم وتوقع سلوك السفر البشري. هذا مهم لأنه يتناول تعقيدات الإدراك البشري واتخاذ القرار في النقل، مما يساعد في تحسين تقييم النظام والتخطيط. من خلال مواءمة وكلاء LLM مع سلوكيات التعلم والتكيف البشرية، يمكن أن يؤدي هذا النهج إلى حلول نقل أكثر فعالية وتجارب مستخدم أفضل.

Un estudio reciente presenta un marco de doble agente que mejora la forma en que los agentes de modelos de lenguaje grande (LLM) pueden ayudar a comprender y predecir el comportamiento de viaje humano. Esto es significativo porque aborda las complejidades de la cognición humana y la toma de decisiones en el transporte, lo que ayuda a una mejor evaluación y planificación del sistema. Al alinear los agentes LLM con los comportamientos de aprendizaje y ajuste humano, este enfoque podría conducir a soluciones de transporte más efectivas y a una mejor experiencia del usuario.

Une étude récente présente un cadre à double agent qui améliore la manière dont les agents de modèle de langage large (LLM) peuvent aider à comprendre et prédire le comportement de voyage humain. Cela est important car cela aborde les complexités de la cognition humaine et de la prise de décision dans les transports, aidant ainsi à une meilleure évaluation et planification des systèmes. En alignant les agents LLM avec les comportements d'apprentissage et d'ajustement humains, cette approche pourrait conduire à des solutions de transport plus efficaces et à une meilleure expérience utilisateur.

A recent study introduces a dual-agent framework that enhances how Large Language Model (LLM) agents can help understand and predict human travel behavior. This is significant because it addresses the complexities of human cognition and decision-making in transportation, ultimately aiding in better system assessment and planning. By aligning LLM agents with human learning and adjustment behaviors, this approach could lead to more effective transportation solutions and improved user experiences.

Aligning LLM agents with human learning and adjustment behavior: a dual agent approach

arXiv:2510.21060v2 Announce Type: replace 
Abstract: Policy optimization (PO) is a cornerstone of modern reinforcement learning (RL), with diverse applications spanning robotics, healthcare, and large language model training. The increasing deployment of PO in sensitive domains, however, raises significant privacy concerns. In this paper, we initiate a theoretical study of differentially private policy optimization, focusing explicitly on its sample complexity. We first formalize an appropriate definition of differential privacy (DP) tailored to PO, addressing the inherent challenges arising from on-policy learning dynamics and the subtlety involved in defining the unit of privacy. We then systematically analyze the sample complexity of widely-used PO algorithms, including policy gradient (PG), natural policy gradient (NPG) and more, under DP constraints and various settings, via a unified framework. Our theoretical results demonstrate that privacy costs can often manifest as lower-order terms in the sample complexity, while also highlighting subtle yet important observations in private PO settings. These offer valuable practical insights for privacy-preserving PO algorithms.

تم نشر دراسة حديثة حول تحسين السياسات بشكل مختلف (DPPO) تركز على تعقيد العينة في تحسين السياسات (PO) في التعلم المعزز (RL). تتناول هذه الدراسة المخاوف المتعلقة بالخصوصية في التطبيقات الحساسة مثل الروبوتات والرعاية الصحية من خلال صياغة تعريف للخصوصية المختلفة يتناسب مع PO وتحليل تعقيد العينة لعدة خوارزميات PO تحت قيود DP.

Se ha publicado un estudio reciente sobre la optimización de políticas diferencialmente privadas (DPPO), centrado en la complejidad de muestra de la optimización de políticas (PO) en el aprendizaje por refuerzo (RL). Esta investigación aborda las preocupaciones de privacidad en aplicaciones sensibles como la robótica y la atención médica al formalizar una definición de privacidad diferencial adaptada a la PO y analizar la complejidad de muestra de varios algoritmos de PO bajo restricciones de DP.

Une étude récente sur l'optimisation de politiques différemment privées (DPPO) a été publiée, se concentrant sur la complexité d'échantillonnage de l'optimisation de politiques (PO) dans l'apprentissage par renforcement (RL). Cette recherche aborde les préoccupations en matière de confidentialité dans des applications sensibles telles que la robotique et les soins de santé en formalisant une définition de la confidentialité différentielle adaptée à la PO et en analysant la complexité d'échantillonnage de divers algorithmes de PO sous des contraintes de DP.

A recent study on differentially private policy optimization (DPPO) has been published, focusing on the sample complexity of policy optimization (PO) in reinforcement learning (RL). This research addresses privacy concerns in sensitive applications such as robotics and healthcare by formalizing a definition of differential privacy tailored to PO and analyzing the sample complexity of various PO algorithms under DP constraints.

On the Sample Complexity of Differentially Private Policy Optimization

One More Thing in AI – Your Shortcut to AI Mastery

Aligning LLM agents with human learning and adjustment behavior: a dual agent approach

Was this article worth reading? Share it

One More Thing in AI

LucidQuery AI

Chattermate

AIvilization

Legion AI

ClassX

Ready to build your own newsroom?