Generative Spatial Intelligence (GSI) Lab

Our research and education focus on computer vision with a particular emphasis on generative models. We develop methods for enhancing generation capabilities, and repurposing these techniques for applications such as content creation and scene understanding.

GSI lab is affiliated with the Department of Computer Science and Engineering at University of California, Merced.

Meet the Team
Lab photo

News

Latest updates from the Generative Spatial Intelligence Lab.

Selected Publications

Recent work from the Generative Spatial Intelligence Lab.

Spatial Reasoning
ICLR 2026
Pursuing Minimal Sufficiency in Spatial Reasoning

Y. Guo, Y. Hou, W. Ma, M. Tang, M. H. Yang

Self-Cross Diffusion
CVPR 2025
Self-Cross Diffusion Guidance for Text-to-Image Synthesis of Similar Subjects

W. Qiu, J. Wang, M. Tang

ReDistill
TMLR 2025
ReDistill: Residual Encoded Distillation for Peak Memory Reduction

F. Chen, G. Datta, M. A. Rafi, H. Jeon, M. Tang

Flow Matching
AAAI 2024
Latent Space Editing in Transformer-Based Flow Matching

V. T. Hu, W. Zhang, M. Tang, P. Mettes, D. Zhao, C. Snoek

Decepticon
IISWC 2023
Decepticon: Attacking Secrets of Transformers

M. Al Rafi, Y. Feng, F. Yao, M. Tang, H. Jeon

Courses

Courses taught at the University of California, Merced.

Advanced Topics in Deep Learning
Advanced Topics in Deep Learning

EECS 242  ·  Graduate  ·  Fall 2025

State-of-the-art deep learning research including generative models, diffusion models, and foundation models.

Details
Introduction to Machine Learning
Introduction to Machine Learning

CSE 176  ·  Undergraduate  ·  Fall 2024

Supervised and unsupervised learning, linear models, neural networks, and practical applications.

Details
Deep Learning
Deep Learning

EECS 230  ·  Graduate  ·  Spring 2024

CNNs, RNNs, Transformers, attention mechanisms, and modern training techniques at scale.

Details
Introduction to Computer Vision
Introduction to Computer Vision

CSE 185  ·  Undergraduate  ·  Fall 2023

Image formation, feature detection, segmentation, object recognition, and deep learning for vision.

Details
Introduction to Deep Learning
Introduction to Deep Learning

CSE 190  ·  Undergraduate  ·  Spring 2026

Neural network fundamentals, convolutional networks, recurrent models, and practical applications.

Details