Undergraduate Student · Tsinghua University

Building language systems that reason, plan, and act with people.

I'm Zixuan (Alex) Wang — a Physics undergraduate at Tsinghua University working on human-centered AI: LLM agents, alignment, and reinforcement learning. Currently a research assistant at Carnegie Mellon University.

↳ the globe is draggable — and this page keeps a few secrets
fig. 1 — a small planet, spinning quietly

I am Zixuan Wang, a Physics undergraduate at Tsinghua University (class of 2023). My research aims toward human-centered AI and autonomous agentic systems — LLMs that reason, plan, and act with people.

I am a remote research assistant at Carnegie Mellon University (ECE), advised by Prof. Andrea Zanette. During my Fall 2025 exchange at UC San Diego, I was an undergraduate researcher at the UCSD MixLab (HDSI) with Dr. Zhen Wang. Earlier, I was an AI intern at MiroMind AI, mentored by Dr. Yuntao Chen.

🌍 Exploration

Traveling and experiencing diverse cultures — including an unforgettable Christmas 2025 by the seaside in La Jolla.

🤖 Competition

Tinker Robot Team, RoboCup@Home 2025 in Salvador, Brazil — an LLM-driven domestic service robot, 8th place globally.

🔬 Visits

Summer 2024 research visits in Singapore: A*STAR, Nanyang Technological University, and the National University of Singapore.

🎷 Music

Saxophone in the Tsinghua University Symphonic Band — performances, festivals, and campus events.

🏐 Sports

Volleyball, swimming, and the gym — ideally all three in one week.

I work on LLM algorithms that enable more reliable, adaptive, and autonomous interaction with humans — leveraging synthetic data, structured supervision, and preference-based training to improve models' understanding of human intent, while ensuring privacy and safety.

Agentic LLM Systems

Long-context reasoning, planning, tool use, and self-refinement in real-world grounded environments.

LLM Alignment

Synthetic data, structured supervision, and preference-based training for human intent — with privacy and safety built in.

RL for LLMs

Training agents for sustained multi-turn interaction and long-horizon task completion.

2023 — present
B.S. in Mathematical and Physical Basic Science (Physics)
Tsinghua University
Minor in Artificial Intelligence
Fall 2025
Exchange Student in Computer Science
UC San Diego
L
Intern · Internship
Looki
2026 — Present · 1 mo ·Beijing, China ·On-site
Human-Computer InteractionLarge Language Models
C
Research Assistant
Carnegie Mellon University — Department of Electrical and Computer Engineering
Feb 2026 — Present · 3 mos ·Remote ·Remote
Agentic AILarge Language Models
M
AI Intern · Internship
Jun 2025 — Jan 2026 · 8 mos ·Shanghai, China ·Hybrid
Agentic AI
S
Research Assistant
Sep 2025 — Dec 2025 · 4 mos ·San Diego, California, United States ·On-site
Human-Computer Interaction
H
Research Assistant
Halıcıoğlu Data Science Institute — UC San Diego (HDSI) — MixLab, advised by Dr. Zhen Wang
Sep 2025 — Dec 2025 · 4 mos ·San Diego, California, United States ·On-site
Large Language ModelsHuman-Centered AI

Papers

Teaser figure for S³-Sim: Simulating Humans for Personalized Language Modeling Submission

S³-Sim: Simulating Humans for Personalized Language Modeling

Jinzhou Tang*, Yufan Zhou*, Zixuan Wang*, Zhaoxiang Feng, Xinle Yu, Steven Ngo, Zhengding Hu, Luoshang Pan, Lianhui Qin, Yufei Ding, Tianmin Shu, Jingbo Shang, Zhiting Hu, Zhen Wang†
In Submission
Teaser figure for Lightweight Neural Refinement for Drift Calibration in Eye Tracking Systems Submission

Lightweight Neural Refinement for Drift Calibration in Eye Tracking Systems

Liu Jiaqi*, Zixuan Wang*, Yuhong Zhang, Dingkang Liang, Jane Hanqi Li, Tzyy-Ping Jung, Gert Cauwenberghs†
In Submission

Technical Reports

Teaser figure for MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents Preprint

Course Projects

Teaser figure for Ego-embodied Reasoner: Egocentric Embodied Reasoning and Planning with MLLM via Reinforcement Learning Course

Ego-embodied Reasoner: Egocentric Embodied Reasoning and Planning with MLLM via Reinforcement Learning

Deep Reinforcement Learning Course Project, Instructed by Professor Huazhe Xu
Teaser figure for Unconditional and Image-conditioned 3D Generation Course

Unconditional and Image-conditioned 3D Generation

3D Visual Computing Course Project, instructed by Professor Li Yi
Teaser figure for Human Skeleton and Skin Generation Course

Human Skeleton and Skin Generation

Fundamentals of Computer Graphics Course Project, the 5-th Jittor AI competition, track 2
Teaser figure for Project Reading Report Course

Project Reading Report

Object Oriented Programming Course Project
2025-02-14

Machine Learning Course Notes - Learning Theory

Comprehensive notes on learning theory, covering PAC learning, VC dimension, and statistical learning foundations.

Machine LearningLearning TheoryCourse Notes
2025-02-14

3D Visual Computing Course Notes

Notes on 3D visual computing, including geometry processing, rendering, and 3D reconstruction techniques.

Computer Graphics3D VisionCourse Notes