Xiaoyuan Yi

Senior Researcher, Microsoft Research Asia, Microsoft Research

Joined

July 2019

Names

Xiaoyuan Yi (Preferred)

Emails

****@microsoft.com (Confirmed)

****@mails.tsinghua.edu.cn (Confirmed)

Personal Links

Career & Education History

Senior Researcher

Microsoft Research Asia, Microsoft Research (research.microsoft.com)

2023 – Present

Researcher

Microsoft Research Asia, Microsoft Research (research.microsoft.com)

2021 – 2023

PhD student

Computer Science and Technology, Tsinghua University, Tsinghua University (tsinghua.edu.cn)

2016 – 2021

Advisors, Relations & Conflicts

Coworker

Yaodong Yang

Present

Social

Guojie Song

Present

Expertise

Responsible AI

value alignment

LLM evaluation

2021 – Present

text generation

poetry generation

2016 – Present

natural language processing

natural language generation

2015 – Present

Publications

Uplifting Human Decision Making in AI Evaluation by Automating Benchmark Validity Analysis
Rodolfo Corona, Sang T. Truong, Ritwik Gupta, Nhi Ngoc Truong, Atnafu Lambebo Tonja, Mena Attia, Fahim Faisal, Kaushal Kumar Maurya, Fred Philippy, Belu Ticona, Sumaya Nur Adan, Fazl Barez, Omar Florez, Supheakmungkol Sarin, Aseem Srivastava, Xiaoyuan Yi, Nick Haber, Dan Klein, Thamar Solorio, Xing Xie et al. (2 additional authors not shown)
- CTB@ICML 2026
- Readers: Everyone
Rethinking Diversity-Preserving RL for Pluralistic Alignment: Empirical Evidence from Rubric-Grounded Moral Reasoning
Zhaowei Zhang, Xiaohan Liu, Xuekai Zhu, Junchao Huang, Ceyao Zhang, Xiang Liu, ZhiYuan Feng, Yaodong Yang, Xiaoyuan Yi, Xing Xie
- Pluralistic-Alignment 2026
- Readers: Everyone
Does LLM Alignment Really Need Diversity? An Empirical Study of Adapting RLVR Methods for Moral Reasoning
Zhaowei Zhang, Xiaohan Liu, Xuekai Zhu, Junchao Huang, Ceyao Zhang, ZhiYuan Feng, Yaodong Yang, Xiaoyuan Yi, Xing Xie
- Submitted to AFAA 2026
- Readers: Everyone
CAReDiO: Enhancing Cultural Alignment of LLM via Representativeness and Distinctiveness Guided Data Optimization
Jing Yao, Xiaoyuan Yi, Jindong Wang, Zhicheng Dou, Xing Xie
- Submitted to ICLR 2026
- Readers: Everyone
AdAEM: An Adaptively and Automated Extensible Measurement of LLMs' Value Difference
Jing Yao, Shitong Duan, Xiaoyuan Yi, Dongkuan Xu, Peng Zhang, Tun Lu, Ning Gu, Zhicheng Dou, Xing Xie
- ICLR 2026 Oral
- Readers: Everyone
MMA-ASIA: A Multilingual and Multimodal Alignment Framework for Culturally Grounded Evaluation
Zheng Weihua, Zhengyuan Liu, Tanmoy Chakraborty, Weiwen Xu, Xiaoxue Gao, Bryan Chen Zhengyu Tan, Bowei Zou, Chang Liu, Yujia Hu, Xing Xie, Xiaoyuan Yi, Jing Yao, Chaojun Wang, Long Li, Rui Liu, Huiyao Liu, Koji Inoue, Ryuichi Sumida, Tatsuya Kawahara, Fan Xu et al. (15 additional authors not shown)
- ICLR 2026 Conference Withdrawn Submission
- Readers: Everyone
Beyond BFI: The CSI for Enhanced Reliability and Validity in Evaluating LLM Personality Traits
Huanhuan Ma, Haisong Gong, Xiaoyuan Yi, Xing Xie, Philip S. Yu, Dongkuan Xu
- Submitted to ICLR 2026
- Readers: Everyone
PICACO: Pluralistic In-Context Value Alignment via Total Correlation Optimization
Han Jiang, Dongyao Zhu, Xiaoyuan Yi, Ziang Xiao, Zhihua Wei, Xing Xie
- Submitted to ICLR 2026
- Readers: Everyone
On the Essence and Prospect: An Investigation of Alignment Approaches for Big Models
Xinpeng Wang, Shitong Duan, Xiaoyuan Yi, Jing Yao, Shanlin Zhou, Zhihua Wei, Peng Zhang, Dongkuan Xu, Maosong Sun, Xing Xie
- IJCAI 2024
- Readers: Everyone
ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generation
Pengcheng Huang, Zhenghao Liu, Yukun Yan, Haiyan Zhao, Xiaoyuan Yi, Hao Chen, Zhiyuan Liu, Maosong Sun, Tong Xiao, Ge Yu, Chenyan Xiong
- NeurIPS 2025 poster
- Readers: Everyone

View all 64 publications

Co-Authors

View all 141 co-authors