Published in arXiv preprint, 2026
This paper introduces SafeTalkCoach, a diversity-driven multi-agent framework for simulating realistic, guideline-grounded parent-teen health conversations.
Published in arXiv preprint, 2025
Presents and evaluates strategies for LLM agents to facilitate knowledge co-construction in asynchronous online discussions.
Published in arXiv preprint, 2025
This paper explores whether LLM agents can simulate customers to evaluate agentic-AI-based shopping assistants.
Published in arXiv preprint, 2025
Introduces the OPeRA dataset for evaluating LLMs on human online shopping behavior simulation.