All publications

arXiv 2026 · Preprint

Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States

Zixuan Wang, Yufan Zhou, Jinzhou Tang, Xinle Yu, Chengjun Wu, Lyumanshan Ye, Zhaoxiang Feng, Letian Peng, Adyasha Patra, Fan Bai, Enze Ma, Zhengding Hu, Jianyang Gu, Zhao Wang, Yufei Ding, Jingbo Shang, Tianmin Shu, Zhiting Hu, Zhen Wang

Overview

Mind2Dialogue generates training conversations using a user simulator and an Oracle assistant that share an evolving simulated mental state. Models learn from the Oracle’s responses while receiving only the user’s persona and conversation history, without access to the simulated state. Evaluation covers personalization and reasoning about beliefs and actions.

Resources