Georgia Tech is a research leader in a wide number of computing, engineering, and scientific disciplines. Explore some of Georgia Tech’s leading published research and learn more about the institute’s long-term vision for advancing education and research progress.


EXPLORE Georgia Tech @ CVPR 2025

Georgia Tech First Authors

CVPR 2025 | Nashville

Profile Picture

Chengyue Huang

PhD, Machine Learning

• Areas of Interest: Machine Learning, Computer Vision, Vision-Language Models

——

• Papers (first author): FRAMES-VQA: Benchmarking Fine-Tuning Robustness across Multi-Modal Shifts in Visual Question Answering

• Program advisor(s): Zsolt Kira

• Fun fact: Second-year part-time PhD student, full-time older sister to two kittens and one dog.

Profile Picture

Bolin Lai

PhD, Machine Learning

• Areas of Interest: Multimodal Learning, Generative Models, Video Understanding

——

• Papers (first author): Unleashing In-context Learning of Autoregressive Models for Few-shot Image Manipulation

• Program advisor(s): James Rehg, Zsolt Kira

• Fun fact: I’m a “senior intern” at Meta (interned at Meta for 3 times). I love soccer, movies and cooking:)

Profile Picture

Fiona Ryan

PhD, Computer Science

• Areas of Interest: Modeling Human Behavior with Computer Vision, Multimodal Learning

——

• Papers (first author): (1) Gaze-LLE: Gaze Target Estimation via Large-Scale Learned Encoders (2) Improving Personalized Search with Regularized Low-Rank Parameter Updates

• Program advisor(s): James Rehg, Judy Hoffman

• Fun fact: Has an undergraduate degree in music.

Profile Picture

Andrew Szot

PhD, Machine Learning

• Areas of Interest: Reinforcement Learning, LLM Agents, Simulation

——

• Papers (first author): From Multimodal LLMs to Generalist Embodied Agents: Methods and Lessons

• Program advisor(s): Zsolt Kira, Dhruv Batra

• Fun fact: I want to develop LLM agents that reason and act to solve complex tasks.

Profile Picture

Lifu Wang

MS, Computer Science

• Areas of Interest: Media and Intelligence

——

• Papers (first author): Scaling Down Text Encoders of Text-to-Image Diffusion Models

• Program advisor(s): Not specified

• Fun fact (aka game guru): In Teamfight Tactics Set 9, I managed to get eight 3-star 5-cost champions in a single game!

Profile Picture

Lexington Whalen

PhD, Computer Science

• Areas of Interest: Robust and Scalable ML/AI

——

• Papers (first author): Early-Bird Diffusion: Investigating and Leveraging Timestep-Aware Early-Bird Tickets in Diffusion Models for Efficient Training

• Program advisor(s): Yingyan (Celine) Lin

• Fun fact: Loves language learning and is fluent in Japanese / is currently studying Mandarin. Big fan of long walks outdoors.

Profile Picture

Haoran You

PhD, Computer Science

• Areas of Interest: Efficient ML Systems, Algorithm-Hardware Co-Design

——

• Papers (first author): Layer- and Timestep-Adaptive Differentiable Token Compression Ratios for Efficient Diffusion Transformers

• Program advisor(s): Yingyan (Celine) Lin

• Fun fact: I see research as a form of cooking — but unlike in the kitchen where I follow safe recipes, in research failure is just a part of the recipe.

Artificial Intelligence

Includes a wide range of areas including natural language processing, data engineering, robotics, and computer vision.

Computer Science Education

Cybersecurity and Privacy

Human-Computer Interaction

ARCHIVE

This site and its content are developed and maintained by the College of Computing’s Office of Communications. The point of contact is Joshua Preston, College Communications Manager.