normxu / build-dev-image Goto Github PK

Hi there 👋🏻,

🔭 I am currently focused on developing cutting-edge multi-modality models capable of generating and understanding text and images natively.

My specific areas of interest include:

• Denoising

• Auto-regressive Generation
• Diffusion

PS: I recognize that all diffusion and next-token generation tasks are inherently denoising tasks. (2024/05)

• Document Understanding & Layout Analysis
• Optical Character Recognition
• Object Detection

PS: Looks like all these tasks can be regarded as next-token generation tasks. (2023/12)

In addition to my current work, I have prior experience in Robotics Perception from my Master's studies. I hope my work can be helpful to you.

Feel free to reach out if you have any questions or if there's anything I can assist with!