I am a 6th-year PhD candidate in the School of Computer Science at Georgia Tech. My research is dedicated to creating efficient and automated ML/AI systems through algorithm-hardware co-design. This involves addressing the challenge posed by the increasing complexity of advanced ML models, such as large language models (LLMs) and large vision models (LVMs), against the constraints of both edge and cloud computing devices.

My research work emphasizes a dual approach: customizing algorithms to maximize their efficiency on specific hardware platforms like GPUs, exemplified by projects such as ShiftAddLLM [arXiv’24] and Linearized-LLM [ICML’24] for efficient LLMs; ShiftAddViT [NeurIPS’23] and Castling-ViT [CVPR’23] for efficient LVMs; and Early-Bird Tickets [ICLR’20 Spotlight] for efficient training and inference of these models. Concurrently, I design customized hardware architectures, such as ASICs or FPGAs, that cooperatively support these algorithms, including ViTCoD [HPCA’23], GCoD [HPCA’22], and EyeCoD [ISCA’22 & IEEE Micro’s Top Pick of 2023]. This integrated strategy ensures cohesive optimization and enhanced performance across diverse computing environments.

My research work has received recognition, including winning First Place at the ACM Student Research Competition (SRC) at ICCAD’23, securing the Best Poster Award at the SCS Poster Competition of Georgia Tech, and being selected as an IEEE Micro’s Top Pick of 2023. Additionally, I was named one of the Machine Learning and Systems Rising Stars of 2023, received the Outstanding Graduate Research Assistant Award, won first place in the University Best Demonstration at DAC’22, and our ViTCoD project was honored with the Meta Faculty Research Award of 2022. More details can be found in my CV.

I am expected to graduate in Spring 2025 and am currently on the job market.


Industry Experience

Selected News

Invited Talks


Conference Reviewer

Journal Reviewer

Volunteering Service