Shuai Xie 😇
Shuai Xie

LLM Inference Engineer

About Me

Shuai Xie is an Algorithm Engineer at JD Explore Academy and the leading member of the Inference Team in the Systems and Foundation Models Department. He is dedicated to the research and development of efficient inference technologies for JD’s foundational large models, as well as the collaborative optimization of training-side technologies. He has led and participated in the development of JD’s self-developed integrated inference deployment framework, which has been applied across various business scenarios within JD.

Download CV
Interests
  • Artificial Intelligence
  • LLM Inference System
Education
  • MSc. Compute Science

    Zhejiang University

  • BSc. Compute Science

    Central South University

📚 My Research

I’m an algorithm and system engineer in the Omninfer team at JD Explore Academy. We apply a range of qualitative and quantitative methods to comprehensively accelerate the inference speed and cut off the cost of large language models, including Post-Training Quantization, Speculative Decoding and Prefill-Decode Disaggregation.

Please reach out to collaborate 😃

Recent Publications
(2024). CoCM: Conditional Cross-Modal Learning for Vision-Language Models. Electronics.
(2024). Cross-Modal Adapter: Parameter-Efficient Transfer Learning Approach for Vision-Language Models. 2024 IEEE International Conference on Multimedia and Expo (ICME).
(2024). Dsdrnet: Disentangling representation and reconstruct network for domain generalization. 2024 International Joint Conference on Neural Networks (IJCNN).
(2024). Generalizing to unseen domains via patchmix. Multimedia Systems.
(2024). Smile: Zero-shot sparse mixture of low-rank experts construction from pre-trained foundation models. arXiv preprint arXiv:2408.10174.
Recent & Upcoming Talks
Recent News

✅ Manage your projects

Easily manage your projects - create ideation mind maps, Gantt charts, todo lists, and more!