#1 in MLSys Challenge
Architected high-performance serving infrastructure, validating global competitiveness in inference and kernel optimization.
About
About
Welcome to Aaron YU's webpage.
01
Bio
Aaron YU is an AI team leader with a Ph.D. in Computer Science and Engineering from The Chinese University of Hong Kong, earned in 2018. He currently leads research and development in foundation models, agent harnesses, and AI productization. His work focuses on large language models (LLMs), vision-language models (VLMs), vision-language-action models (VLAs), inference systems for models with hundreds of billions of parameters, and coding-agent infrastructure.
02
News
Commercial scale
Scaled multimodal AI architectures built on 2B+ parameter VLMs to drive RMB 50M+ in annual revenue.
Global leaderboard #1
Led post-training teams that achieved the top ranking on MMBench in April 2024.
Research
Published key work on parallel MTP decoding at ACL Findings and graph learning in Neural Networks.
Global leaderboard #1
Achieved the top ranking on C-Eval in September 2023 with 100B+ parameter LLMs developed for public-sector deployment.
Wu Wenjun AI Science and Technology Progress Award
Received First Prize for contributions to applied artificial intelligence.