Yuxiang Xu (许宇翔)
PhD Candidate in AI for Science
Institute of Mechanics, Chinese Academy of Sciences
University of Chinese Academy of Sciences
Institute of Mechanics, Chinese Academy of Sciences
University of Chinese Academy of Sciences
I am a PhD candidate specializing in AI for Science, with a core focus on the application of Large Language Models (LLMs) to scientific challenges. My research interests lie at the intersection of physics and artificial intelligence, specifically exploring the emerging capabilities of Large Language Models (LLMs) and Vision-Language Models (VLMs) agents in scientific discovery.
Chinese Physics Letters, 2025
Physical Review D, 2025
arXiv preprint, 2025
An advanced framework exploring the capabilities of Vision-Language Models (VLMs). Cinego focuses on enhancing multimodal understanding, designed to tackle complex visual reasoning tasks with improved efficiency.
The next iteration of the Cinego architecture, integrating Reasoning-based optimizations (R1). This project aims to push the boundaries of LLM/VLM post-training for scientific agents and complex decision-making scenarios.