SciFig: Towards Automating Scientific Figure Generation

TL;DR

SciFig is an AI system that automates the creation of scientific figures from research texts, improving efficiency and quality through hierarchical layout strategies and iterative reasoning.

Contribution

The paper introduces SciFig, a novel AI pipeline that automates scientific figure generation using hierarchical layout parsing and iterative feedback, with a new evaluation framework.

Findings

Achieves 70.1% overall quality on dataset-level evaluation

Consistently high scores in visual clarity and scientific accuracy

Open-sourced figure generation pipeline and evaluation benchmark

Abstract

Creating high-quality figures and visualizations for scientific papers is a time-consuming task that requires both deep domain knowledge and professional design skills. Despite over 2.5 million scientific papers published annually, the figure generation process remains largely manual. We introduce $\textbf{SciFig}$, an end-to-end AI agent system that generates publication-ready pipeline figures directly from research paper texts. SciFig uses a hierarchical layout generation strategy, which parses research descriptions to identify component relationships, groups related elements into functional modules, and generates inter-module connections to establish visual organization. Furthermore, an iterative chain-of-thought (CoT) feedback mechanism progressively improves layouts through multiple rounds of visual analysis and reasoning. We introduce a rubric-based evaluation framework that analyzes 2,219 real scientific figures to extract evaluation rubrics and automatically generates comprehensive evaluation criteria. SciFig demonstrates remarkable performance: achieving 70.1$\%$ overall quality on dataset-level evaluation and 66.2$\%$ on paper-specific evaluation, and consistently high scores across metrics such as visual clarity, structural organization, and scientific accuracy. SciFig figure generation pipeline and our evaluation benchmark will be open-sourced.