Zheyuan Zhang

I am a first-year Ph.D. student at CMU LTI, advised by Prof. Sherry Wu at the WInE Lab🍷.

I did my master at THUKEG where I was fortunate to work with Prof. Juanzi Li and Prof. Zhiyuan Liu. I visited CLIP Lab@UMD in 2024 where I work with Prof. Jordan Boyd-Graber.

I obtained my Bachelor of Laws (LL.B.) degree in Philosophy, Politics, and Economy (PPE) and a minor in Psychology at Tsinghua University.

I have a broad interest in social sciences and cognitive sciences, receiving research training in these interdisciplinary fields. You can find my CV here.

Research

I am dedicated to understanding advanced models from the perspectives of intelligence and cognition, and to developing AI systems and applications that truly benefit humanity and society. Some projects I worked on are listed below:

Cognitively Understanding and Improving Language Models: Cognitive Diagnostic of LLMs, EmoBench, Arithmetic Learning, KoLA.
AI for Education: SimClass, MAIC, Slide2Lecture, LittleMu.

I’m currently working on Human-Agent Interaction, particularly Deep Research, Collaborative Literature Exploration, and Sensemaking.

I believe deep research agents should be fundamentally user-centric—assisting users to make sense and inspiring new ideas—making sustained user engagement essential. I study how users make decisions (data) and how to sensemake collaboratively (model).

News

[Dec 2025] MAIC has reached more than 90,000 learners.
[Aug 2025] Excited to join CMU LTI as a Ph.D. student!
[Jan 2025] Our paper SimClass is accepted by NAACL 2025. See you in New Mexico!
[Nov 2024] Our paper Slide2Lecture is accepted by KDD 2025!!!
[Sept 2024] I’m honored to be selected as a Siebel Scholar of Class 2025.
[July 2024] Our online teaching platform MAIC was deployed at Tsinghua University as the first multi-agent AI education system, offering two courses to over 500 students in the previous semester. Check related works here: SimClass, Slide2Lecture.

Selected Publications

* indicates equal contribution. Please refer to my Google Scholar profile for the full list.

Cognitively Understanding and Improving Language Models:

1. NAVIG: Natural Language-guided Analysis with Vision Language Models for Image Geo-localization
Zheyuan Zhang, Runze Li, Tasnim Kabir, Jordan Boyd-Graber
arXiv preprint, 2025

2. Exploring the Cognitive Knowledge Structure of Large Language Models: An Educational Diagnostic Assessment Approach
Zheyuan Zhang, Jifan Yu, Juanzi Li, Lei Hou
Findings of EMNLP 2023

3. EmoBench: Evaluating the Emotional Intelligence of Large Language Models
Sahand Sabour, Siyang Liu, Zheyuan Zhang, June M. Liu, Jinfeng Zhou, et al.
ACL 2024

4. Reverse That Number! Decoding Order Matters in Arithmetic Learning
Daniel Zhang-Li, Nianyi Lin, Jifan Yu, Zheyuan Zhang, Zijun Yao, et al.
arXiv preprint, 2024

AI for Education

5. Simulating Classroom Education with LLM-Empowered Agents
Zheyuan Zhang*, Daniel Zhang-Li*, Jifan Yu, Linlu Gong, Jinchang Zhou, Zhiyuan Liu, Lei Hou, Juanzi Li
NAACL 2025

6. LittleMu: Deploying an Online Virtual Teaching Assistant via Heterogeneous Sources Integration and Chain of Teach Prompts
Shangqing Tu*, Zheyuan Zhang*, Jifan Yu, Chunyang Li, Siyu Zhang, Zijun Yao, et al.
CIKM 2023

7. Awaking the Slides: A Tuning-free and Knowledge-regulated AI Tutoring System via Language Model Coordination
Daniel Zhang-Li*, Zheyuan Zhang*, Jifan Yu, Joy Lim Jia Yin, Shangqing Tu, Linlu Gong, et al.
KDD 2025

8. From MOOC to MAIC: Reshaping Online Teaching and Learning through LLM-driven Agents
Jifan Yu, Zheyuan Zhang, Daniel Zhang-Li, Shangqing Tu, Zhanxin Hao, Ruimiao Li, et al.
arXiv preprint, 2024

Professional Services

Reviewer: CIKM 2024, ACL/NAACL/EMNLP 2024/2025 (ARR), AAAI 2025, KDD 2025/2026, WWW 2026.

Awards / Honors

[2025] Outstanding Master’s Thesis. Title: Intelligent Classrooms based on Large Language Model Agents.
[2025] Outstanding Graduates. Outstanding Graduates of Beijing. Top 2 in Master Students.
[2024] Siebel Scholar, Class 2025. Top 5 in Tsinghua University, 78 students around the world in Class 2025.
[2024] Huiyan Talent Comprehensive Scholarship. Comprehensive Scholarship for guaduate students.
[2022] Outstanding Graduates. Selected as Outstanding Graduates of Tsinghua University, Outstanding Graduates of Beijing.
[2021] Toyota Scholarship of Tsinghua University. One of the best Comprehensive Scholarships in Tsinghua University.
[2019, 2020, 2021] Excellent Comprehensive Scholarship of Tsinghua Unversity. Awards for comprehensive performance in academic and others.
[2019, 2020, 2021] Excellent Scholarship of Tsinghua Unversity. Exceptional performance in Academic (2019, 2020, 2021); Social Work (2020); and Sports (2019, 2020, 2021).

Misc

When I’m not doing research, I’m passionate about sports, especially basketball and track and field. I hold the standing long jump record at Tsinghua University (3.22m). I love music, reading, and movies, which keep me thoughtful, enthusiastic, and fulfilling.