Sheng-Yu Wang

Title: Innovator Sheng-Yu Wang: Pioneering Data Attribution in Diffusion Models

Introduction

Sheng-Yu Wang is a notable inventor based in Pittsburgh, PA (US). He has made significant contributions to the field of machine learning and image generation. His innovative work focuses on data attribution for diffusion models, which has implications for various applications in artificial intelligence.

Latest Patents

Wang holds a patent titled "Data attribution for diffusion models." This patent describes a method for performing training attribution by identifying a synthesized image and a training image. The synthesized image is generated by an image generation model trained with the training image. A machine learning model computes first attribution features for the synthesized image using a first mapping layer and second attribution features for the training image using a different mapping layer. An attribution score is then generated based on these features, indicating the degree of influence the training image has on generating the synthesized image. Wang has 1 patent to his name.

Career Highlights

Sheng-Yu Wang is currently employed at Adobe, Inc., where he continues to push the boundaries of technology and innovation. His work at Adobe has allowed him to collaborate with other talented individuals in the field, enhancing the quality and impact of his contributions.

Collaborations

One of Wang's notable collaborators is Alexei A Efros, a respected figure in the field of computer vision and machine learning. Their partnership exemplifies the collaborative spirit that drives innovation in technology.

Conclusion

Sheng-Yu Wang's work in data attribution for diffusion models represents a significant advancement in machine learning. His contributions are paving the way for future innovations in image generation and artificial intelligence.