Xingyi Yang

🚀👨‍💻🥁💡Hi, my name is Xingyi Yang (Adam)💡🥁👨‍💻🚀. I am a fourth-year Ph.D student at National University of Singapore(NUS), affiliated with the Learning and Vision Lab. I am working under the supervision of Prof.Xinchao Wang.

I am now a visiting PhD at University of Oxford, advised by Prof.Philip Torr.

My current research interest lies in Deep Model Reuse and its Applications:

• Deep Model Reuse: Rather than trying to cook up a universal model in one grand kitchen experiment, my vision is to reuse pre-trained neural networks to do new things. By composing their expertise, editing their behaviour and enhancing their reliability, the goal is to craft models that become Jacks of all trades, or at least, one step closer to the dream of Artificial General Intelligence (AGI).
• Compositionality & Modularity: Building ML system is like playing a game of LEGO, where I piece together elements from various tasks, concepts, and logic, crafting cost-effective AI masterpieces.
• Generative Models and 3D Vision: I’m super excited about generative models, especially diffusion-based ones. They have the power to create magic in 3D and even 4D worlds – all, hopefully, with a side of solid theory.

CV

I believe in Slow Science