Shengkun (Bryson) Tang Welcome to my website~ My name is Shengkun Tang. You can call me Bryson for short. Currently, I am a research intern in Alibaba Qwen Team. Besides, I am a PhD student of Machine Learning in MBZUAI, under the supervision of Prof. Zhiqiang Shen. During my gap year, I had a wonderful time as an research assistant in DASLab in ISTA , working with Prof. Dan Alistarh. Besides, I had close collaboration with Prof. Dongkuan Xu (NCSU) and Dr. Yaqing Wang (Google DeepMind), working on efficent multi-modal models. I finished B.E. in Remote Sensing at Wuhan University , under the supervision of Prof. Jian Yao and Prof. Xin Su. Email  /  Google Scholar  /  Github Last updated: June 27th 2025

My research interests lie on Landable Artificial Intelligence, focusing on the Resource Efficiency and Trustworthy of AI System. My research covers the whole pipeline of AI system, providing full-stack solutions from theoretical optimization methods and data-centric strategies to the development of efficient, interpretable and reliable deep learning techniques and the co-design of algorithms and hardware.

• Resource-Efficient Training & Inference Algorithms

• Data Optimization to Improve Data Quality & Efficiency

• Scalable Methods for AI Systems with Theoretical Guarantees

• Algorithm-Hardware Co-design for Acceleration

• Application Scenario: Multi-Modal (Vision-Language), Uni-Modal (NLP, Computer Vision)

	MosaicDiff: Training-free Structural Pruning for Diffusion Model Acceleration Reflecting Pretraining Dynamics Bowei Guo, Shengkun Tang, Cong Zeng, Zhiqiang Shen [ICCV 2025] International Conference on Computer Vision, ICCV 2025 Paper
	Do Large Language Models Perceive Orderly Number Concepts as Human? Xuanjie Liu, Cong Zeng, Shengkun Tang, Ziyu Wang, Gus Xia [Re-Align Workshop, ICLR 2025] 2nd Workshop on Representational Alignment, ICLR 2025 Paper
	DALD: Improving Logits-based Detector without Logits from Black-box LLM Cong Zeng*, Shengkun Tang*, Xianjun Yang, Yuanzhou Chen, Yiyou Sun, Yao Li, Haifeng Chen, Wei Cheng, Dongkuan Xu [NeurIPS 2024] The Thirty-eighth Annual Conference on Neural Information Processing Systems arXiv / code
	Adadiff: Accelerating diffusion models through step-wise adaptive computation Shengkun Tang, Yaqing Wang, Caiwen Ding, Yi Liang, Yao Li, Dongkuan Xu [ECCV 2024] European Conference on Computer Vision arXiv / code
	You Need Multiple Exiting: Dynamic Early Exiting for Accelerating Unified Vision Language Model Shengkun Tang, Yaqing Wang, Zhenglun Kong, Tianchi Zhang, Yao Li, Caiwen Ding, Yanzhi Wang, Yi Liang, Dongkuan Xu [CVPR 2023] The IEEE/CVF Conference on Computer Vision and Pattern Recognition arXiv / code
	DDR-Net: Learning Multi-Stage Multi-View Stereo With Dynamic Depth Range Puyuan Yi*, Shengkun Tang*, Jian Yao Preprint, 2021 arXiv / code
	Scale-robust deep-supervision network for mapping building footprints from high-resolution remote sensing images Haonan Guo, Xin Su, Shengkun Tang, Bo Du, Liangpei Zhang IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021 PDF

	Qwen Team, Alibaba, 06/2025 - Now Research Intern Mentor: Bo Zheng and Dayiheng Liu
	SenseTime, Engineering & Intelligent Automotive Group (IAG), 06/2021 - 10/2021 & 05/2022 - 07/2023 Vision Algorithm Intern; System Developer Project: SenseRobot Chess Robotic, working with Ruodai Li Project: Large-Scale Self-Driving System Development

Baidu Astar Developer Competition, 05/2020 - 10/2020 Ranking: 7/2305 (teams)