About
I am an autonomous driving algorithm engineer at NIO, primarily responsible for the research and development of
AI Engine, AI Infra, and HPC (High Performance Computing) etc.
In addition, I am a co-founder of Tsingpe Intelligence Inc., a start-up founded by alumni of Tsinghua University and Peking University.
Within this AI+CAX focused company, I serve as Director of Product and R&D, overseeing the engineering and practical deployment of AI algorithms.
Prior to this, I obtained my Bachelor’s degree in Electronic Engineering (EE) from Ocean University of China in 2020,
followed by a Master’s degree in Computer Science (CS) in 2023.
Research
My work centres on areas such as Deep Neural Network Compression, LLM Knowledge Distillation, High Performance Computing,
Efficient Inference, Efficient Training, AI Engine, AI infra, and other R&D arounding CUDA.
I am also engaged in the design and optimisation of autonomous-driving-oriented algorithms, such as World Models.
Publications
- X Liu, LN Wang, W Liu, G Zhong "Incremental layers resection: a novel method to compress neural networks", IEEE Access 2019. [PDF]
- X Zhang, H Zeng, X Liu, Z Yu, H Zheng, B Zheng “In situ holothurian noncontact counting system: A general framework for holothurian counting”, IEEE Access 2020. [PDF]
- G Zhong, W Liu, H Yao, T Li, J Sun, X Liu “Merging similar neurons for deep networks compression”, Cognitive Computation 2020. [PDF]
- LN Wang, W Liu, X Liu, G Zhong, PP Roy, J Dong, K Huang “Compressing deep networks by neuron agglomerative clustering”, Sensors 2020. [PDF]
- X Liu, W Liu, LN Wang, G Zhong “Deep architecture compression with automatic clustering of similar neurons”, PRCV 2021. [PDF]
- Z Ding, X Liu, G Zhong, D Wang “Steelygan: semantic unsupervised symbolic music genre transfer”, PRCV 2022. [PDF]
- 刘翔，祝静，仲国强，顾永健 等 "量子原型聚类", 计算机科学 2023. [DOI]
Contact
Email: ailven.x.liu@gmail.com