Graduate Student in Computer Science · Georgia Tech
I am a first-year graduate student in Computer Science at the Georgia Institute of Technology, where I am advised by Prof. Humphrey Shi. I received my undergraduate degree from the Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University (also known as the Yao Class, directed by Turing Award laureate Andrew Chi-Chih Yao). Previously, I interned at New York University, where I was advised by Prof. Saining Xie.
Humans exhibit an extraordinary capacity to integrate information from multiple sensory modalities, such as vision, auditory, and tactile inputs, to navigate and interpret their environments with remarkable efficiency. This multimodal integration leverages the complementary strengths of each sensory channel, facilitating a coherent and comprehensive understanding of complex surroundings. Inspired by this cognitive prowess, my long-term objective is to develop AI systems that emulate human-like multimodal synthesis, thereby enhancing robustness and adaptability in both generative and understanding tasks.
Specifically, I focus on multimodal generative models, ranging from text-to-image synthesis to Multimodal Large Language Models (MLLMs) and unified models.
I joined Adobe as a research intern.
Our papers DuoGen, DIG, SAGE, and PAI-Bench were accepted by CVPR 2026.
I joined Georgia Tech as a graduate student in Computer Science.
I graduated from Tsinghua University.
Our paper Science-T2I was accepted by CVPR 2025.
I joined MSRA for an internship with Dr. Bin Li.
Our paper CHOPS was accepted by COLM 2024.
I started an internship at NYU Courant in New York, advised by Prof. Saining Xie.
(* for equal contribution)
arXiv 2026
WorldBagel is a unified Vision-Language-Action-World (VLAW) framework built on BAGEL that systematically investigates the role of unification in world modeling. On LIBERO, Language Table, and Franka, it consistently outperforms task-specific alternatives and learns action representations that are more structured and semantically aligned with visual and linguistic context.
CVPR 2026 Highlight
DuoGen enables general-purpose interleaved multimodal generation by fusing a pretrained MLLM with a video-pretrained DiT. Leveraging a curated 298k-sample dataset, DuoGen significantly outperforms open-source baselines in visual fidelity and consistency and achieves performance comparable to Nano-banana.
CVPR 2026
DIG optimizes long-form video understanding by distinguishing between "global" and "localized" queries, applying different strategies for each with a training-free approach.
CVPR 2026
SAGE introduces a human-inspired agentic framework that replaces resource-heavy frame processing with iterative reasoning and a diverse toolkit for efficient long video understanding by leveraging a novel synthetic data pipeline and a specialized RL strategy.
CVPR 2026 Oral
PAI-Bench is a new benchmark designed to evaluate Physical AI capabilities across video generation and understanding. The study finds that while current models produce high-quality visuals, they lack the physical common sense and reasoning required to truly understand real-world dynamics.
CVPR 2025
We introduce SciScore, a reward model that improves generative models' scientific accuracy. Trained in two stages, it achieves human-level performance in evaluating scientific image realism on Science-T2I dataset.
COLM 2024
CHOPS uses small and large LLMs to create an efficient, safe LLM agent that accesses user data, interacts with systems, and delivers accurate responses.
2025.08 - Present
Graduate Student
Research Advisor: Prof. Humphrey Shi
Outstanding Scholarship in Social Work from Tsinghua University.
Mr. and Mrs. Wong Yi-Chung Award, Friends of Tsinghua University.
Outstanding Scholarship in Social Work from Tsinghua University.
Member of the Chinese team of the 21st Asian Physics Olympiad (APhO).
Gold Medalist in the 37th Chinese Physics Olympiad (CPhO), ranking tenth nationwide.
