Hanze Guo

Hanze Guo

Ph.D. student working on LLM alignment, personalized reasoning, retrieval, and recommendation.

Hanze Guo

I am a Ph.D. student in Artificial Intelligence at the Gaoling School of Artificial Intelligence, Renmin University of China. My research focuses on large language model alignment, personalized reasoning, information retrieval, and recommendation systems.

My recent work studies how to make language models more controllable, personalized, and explainable. I have published three first-author CCF-A papers, including one NeurIPS paper, and have additional CCF-A submissions under review.

Research Interests

  • Large language model alignment and preference modeling
  • Personalized reasoning, controllable generation, and model distillation
  • Information retrieval, recommendation systems, and explainable recommendation
  • Multimodal search agents and LLM applications

Selected Publications

Counterfactual Reasoning for Steerable Pluralistic Value Alignment of Large Language Models

NeurIPS 2025, CCF-A, Poster
Hanze Guo, Jing Yao, Xiao Zhou, Xiaoyuan Yi, Xing Xie
arXiv

We propose COUPLE, a counterfactual reasoning framework for steerable pluralistic value alignment of large language models. The framework uses structural causal models and counterfactual sampling to support fine-grained control over complex human value preferences.

SoREX: Towards Self-Explainable Social Recommendation with Relevant Ego-Path Extraction

ACM Transactions on Information Systems (TOIS), 2025, CCF-A
Hanze Guo, Yijun Ma, Xiao Zhou
ACM DL | Code

We develop SoREX, a self-explainable GNN-based social recommendation framework that extracts factor-specific ego-paths as human-understandable explanations while maintaining strong recommendation accuracy.

Why not Collaborative Filtering in Dual View? Bridging Sparse and Dense Models

ACM Transactions on Information Systems (TOIS), 2026, CCF-A
Hanze Guo, Jianxun Lian, Xiao Zhou

We study how sparse user-item interaction structures can be integrated with dense recommendation models, and propose a plug-and-play dual-view framework that improves performance, especially for long-tail items.

Education

Renmin University of China, Gaoling School of Artificial Intelligence
Ph.D. student in Artificial Intelligence, Sept. 2022 - Present

University of Electronic Science and Technology of China
B.S. in Computer Science and Technology, Sept. 2018 - Jun. 2022
GPA: 3.96/4.0; Outstanding Graduate of Sichuan Province