Reproduce “Content-Edited Layout Generation for Graphic Design” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Shirong Yang, Zheyi Pan, Ying Cao
CELG adapts a supplied image and generates a compatible layout for a design intention and optional foreground requirements. It uses a pretrained image editor and layout generator that exchange information during synchronized generation. On the InstructEdit test split, joint generation reaches 82.4% C-Sat. and TemplateFID 62.73, versus 76.9% and 66.71 for Edit→Layout. This constructed test does not establish real-world performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.