Skip to content

Source paper

AutoFigure-Edit: Generating Editable Scientific Illustration

Zhen Lin, Qiujie Xie, Minjun Zhu and 10 more

Zhen Lin, Qiujie Xie, Minjun Zhu, Shichen Li, Qiyao Sun, Enhao Gu, Yiran Ding, Ke Sun, Fang Guo, Panzhong Lu, Zhiyuan Ning, Yixuan Weng, Yue Zhang

arXiv 2603.06674 · 202617 upvotes on Hugging Face

Research ideas from this paper (3)

Generated by AI. The paper’s authors did not write these.

AI summary

This paper introduces AUTOFIGURE-EDIT, an end-to-end automated system designed to generate fully editable, high-quality scientific illustrations from long-form scientific texts. The main contribution is a framework that explicitly decouples layout planning from visual rendering by generating structural SVG scaffolds, allowing users to apply reference-guided style control and perform fine-grained interactive edits on the resulting diagrams.

How it works (AI summary)

The system uses a sequential multi-stage pipeline: (1) generating a style-conditioned raster draft from text, (2) applying instance segmentation to extract an indexed structural layout, (3) extracting transparent visual assets, (4) using a Vision-Language Model to generate and refine an SVG layout template, and (5) injecting the raster assets into the SVG template to produce fully editable vector graphics.