Hi, I'm Dachuan.

/DAH-chwahn/

A second-year CS Ph.D. student at Georgia Tech, working with Prof. Wenke Lee. Previously, I received my M.S. and B.S. in CS from Tsinghua University.

Research

I have been working on LLMs/MLLMs reasoning and inference, with a focus on efficiency and alignment.

SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs Dachuan Shi, Abedelkadir Asi, Keying Li, Xiangchi Yuan, Leyan Pan, Wenke Lee, Wen Xiao
Preprint, 2025
[Paper] [Code] [Website]

LaCache: Ladder-Shaped KV Caching for Efficient Long-Context Modeling of Large Language Models Dachuan Shi, Yonggan Fu, Xiangchi Yuan, Zhongzhi Yu, Haoran You, Sixu Li, Xin Dong, Jan Kautz, Pavlo Molchanov, Yingyan Celine Lin
ICML, 2025
[Paper] [Code]

CrossGET: Cross-Guided Ensemble of Tokens for Accelerating Vision-Language Transformers Dachuan Shi, Chaofan Tao, Anyi Rao, Zhendong Yang, Chun Yuan, Jiaqi Wang
ICML, 2024
[Paper] [Code]

UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers Dachuan Shi, Chaofan Tao, Ying Jin, Zhendong Yang, Chun Yuan, Jiaqi Wang
ICML, 2023
[Paper] [Code] [Website]

Mitigating Forgetting Between Supervised and Reinforcement Learning Yields Stronger Reasoners Xiangchi Yuan, Xiang Chen, Tong Yu, Dachuan Shi, Can Jin, Wenke Lee, Saayan Mitra
Preprint, 2025
[Paper]

Superficial Self-Improved Reasoners Benefit from Model Merging Xiangchi Yuan, Chunhui Zhang, Zheyuan Liu, Dachuan Shi, Leyan Pan, Soroush Vosoughi, Wenke Lee
EMNLP, 2025
[Paper] [Code]

Supervised Fine-tuning in turn Improves Visual Foundation Models Xiaohu Jiang, Yixiao Ge, Yuying Ge, Dachuan Shi, Chun Yuan, Ying Shan
Preprint, 2024
[Paper] [Code]

Masked Generative Distillation Zhendong Yang, Zhe Li, Mingqi Shao, Dachuan Shi, Zehuan Yuan, Chun Yuan
ECCV, 2022
[Paper] [Code]

Experience

Honors

Services

TA @ CS4476 Computer Vision, Georgia Tech, Spring 2025
Reviewer @ ICML 20-[24,25], NeurIPS 20-[23,24,25], ICLR 20-[24,25], CVPR 20-[24,25], ICCV 2025, ECCV 2024, ACL ARR 2025, and TPAMI