Hello there, Kun Wang here.
Incoming CS PhD Student @ Princeton University | Mathematics & CS Undergrad @ UC San Diego

Currently, I am a research assistant advised by Prof. Rose Yu and Prof. Yian Ma in the Rose Spatio-Temporal Lab. In the past, I have worked with Prof. Manmohan Chandraker on 3D reasoning ability of Large Language Models.

My current research interest lies within the parallel of human and AI agents:

• 👀 Cognition: How humans and AI process complex environments and senses.
• 🧠 Reasoning: Problem-solving from sequential planning to parallel processing.
• 📊 Modeling: Modeling and guiding LLM behavior via statistical methods.
• 💡 Creativity: The synergy of human insight and algorithmic processes.

Works:

* denotes equal contribution; # denotes the corresponding author.

Discovering Latent Structural Causal Models from Spatio-Temporal Data
Kun Wang*, Sumanth Varambally*, Duncan Watson-Parris, Yian Ma, Rose Yu^#
International Conference on Machine Learning (ICML), 2025
Oral Presentation, NeurIPS 2024 Causal Representation Learning Workshop
[Paper]

SceneProg: Program Synthesis for 3D Scene Generation using LLMs
Kunal Gupta, Ishit Mehta, Kun Wang, Nicholas Chua, Yan Deng, Abhimanyu Krishna, Ravi Ramamoorthi, Manmohan Chandraker^#
Under review, 2025