Dong Li

Dong Li is a Director at AMD AIG-Models Team. He leads a team for AI algorithm optimization covering diverse applications (e.g., efficient LLM/VLM/image/video generation) on broad AMD devices (GPU, NPU, CPU) and customer support. Prior to joining AMD/Xilinx, he was a researcher at a Chinese AI startup for computer vision project research and landing. Dong received his Bachelor’s degree and PhD from Tsinghua University in 2012 and 2017 respectively, and served as a visiting scholar at University of California, Merced in 2015-2016. He has 10+ years AI/ML algorithm experience with 30+ publications on top-tier AI conferences (e.g., NeurIPS, CVPR). He won 1st place in fidelity track of NTIRE2022 Efficient Image Super-resolution Challenge, and Team Player Award in 2022 at AMDr.

Posts by Dong Li

Introducing AMD EVLM: Efficient Vision-Language Models with Parameter-Space Visual Conditioning

A novel approach that replaces visual tokens with perception-conditioned weights, reducing compute while maintaining strong vision-language performance.

August 22, 2025 by Zhenhua Liu, Xuanwu Yin, Dong Li, Emad Barsoum

GEAK: Introducing Triton Kernel AI Agent & Evaluation Benchmarks

AMD introduces GEAK, an AI agent for generating optimized Triton GPU kernels, achieving up to 63% accuracy and up to 2.59× speedups on MI300X GPUs.

August 01, 2025 by Jianghui Wang, Vinay Joshi, Saptarshi Majumder, Chao Xu, Bin Ding, Ziqiong Liu, Pratik Prabhanjan Brahma, Dong Li, Zicheng Liu, Emad Barsoum