Haoxuan Sun
M.S. in Computer Science, Vanderbilt University
I am a first-year master's student at Vanderbilt University supervised by Prof. Soheil Kolouri for my master's thesis. My research focuses on world models — understanding how to model dynamic world evolution and generate interactive environments. I am broadly interested in generative models, reinforcement learning, and procedural content generation for games. I am actively seeking PhD positions starting Fall 2027.
Research Interests
World Models
Dynamic world evolution modeling and interactive environment generation for games and simulations.
Generative Models
Diffusion models, VAEs, and their applications to content generation and sequential authoring.
Procedural Content Generation
AI-driven game level and world generation conditioned on player behavior and designer constraints.
Reinforcement Learning
RL-based approaches for controllable generation and agent-environment interaction.
Selected Projects
WorldScribe
Framing game world generation as online sequential authoring conditioned on player behavioral trajectories and designer constraints. Unlike static one-shot generation approaches, WorldScribe models the dynamic co-evolution of game worlds and player interactions over time.
AIGC Detector Robustness
Investigated the robustness of AI-generated content detectors under various adversarial perturbations and distribution shifts, revealing vulnerabilities in existing detection pipelines.
Cross-Modal Signal Generation (PPG-to-ECG)
Developed generative models for cross-modal physiological signal translation, converting photoplethysmography (PPG) signals to electrocardiogram (ECG) waveforms.
Publications
Papers in preparation. Check back soon.
Education
M.S. in Computer Science
Vanderbilt University
2025 – Present
B.S. in Artificial Intelligence
Shanghai Jiao Tong University(SJTU)
2020 – 2024