2 papers
cs.CV2026
Editable Visual Design
Junyan Ye, Wei Liu, Dongzhi Jiang +9
While diffusion base models such as GPT-Image-2 and Nano-Banana exhibit remarkable visual expressiveness, their end-to-end generation inherently yields flattened bitmaps with error…
cs.CV2026
Open-Source Image Editing Models Are Zero-Shot Vision Learners
Wei Liu, Jiaxin Lin, Rui Chen
Recent studies have shown that large generative models can solve vision tasks they were not explicitly trained for. However, existing evidence relies on closed-source models~(Veo~3…