Jing Shi is a research scientist at Adobe Research. His primary interests are in visual perception and generation/manipulation with the interaction of language. His recent work has focused on language-based image editing, scene understanding, and content authenticity. He also has broader interests in the principled way to understand representation learning, reinforcement learning, etc.

Before joining Adobe, he obtained CS Ph.D. at the University of Rochester in 2022 and B.E. degree at the University of Electronic Science and Technology of China.

For more information, please visit his personal webpage.

Publications

FineMatch: Aspect-based Fine-grained Image and Text Mismatch Detection and Correction

Hua, H., Shi, J., Kafle, K., Jenni, S., Zhang, D., Collomosse, J., Cohen, S., Liu, J. (Oct. 1, 2024)

European Conference on Computer Vision (ECCV)

InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning

Shi, J., Xiong, W., Lin, Z., Jung, H. (Jun. 17, 2024)

Conference on Computer Vision and Pattern Recognition (CVPR)

VIXEN: Visual Text Comparison Network for Image Difference Captioning

Black, A., Shi, J., Fan, Y., Bui, T., Collomosse, J. (Jan. 14, 2024)

AAAI Conference on Artificial Intelligence (AAAI)