Automatic Method Illustration Generation for AI Scientific Papers via Drawing Middleware Creation, Evolution, and Orchestration

TL;DR

This paper introduces FigAgent, a multi-agent framework that automates the creation of scientific method illustrations by mimicking human drawing practices, using reusable middlewares and an explore-and-select strategy.

Contribution

The paper presents a novel multi-agent system with reusable drawing middlewares and an explore-and-select strategy for automatic, high-quality method illustration generation.

Findings

Effective in generating complex scientific illustrations

Outperforms baseline methods in quality and adaptability

Demonstrates robustness across diverse MI styles

Abstract

Method illustrations (MIs) play a crucial role in conveying the core ideas of scientific papers, yet their generation remains a labor-intensive process. Here, we take inspiration from human authors' drawing practices and correspondingly propose \textbf{FigAgent}, a novel multi-agent framework for high-quality automatic MI generation. Our FigAgent distills drawing experiences from similar components across MIs and encapsulates them into reusable drawing middlewares that can be orchestrated for MI generation, while evolving these middlewares to adapt to dynamically evolving drawing requirements. Besides, a novel Explore-and-Select drawing strategy is introduced to mimic the human-like trial-and-error manner for gradually constructing MIs with complex structures. Extensive experiments show the efficacy of our method.