AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations [ICLR 2026]

From Text to Publication-Ready Diagrams
AutoFigure is an intelligent system that leverages Large Language Models (LLMs) with iterative refinement to generate high-quality scientific figures from text descriptions or research papers.

Quick Start • Web Interface • Configuration • API Reference

https://github.com/user-attachments/assets/d0c954a9-9cf3-4c8b-8b04-71d75a68854c

🔥 News

• [2026.03.24] 🧠 Our sister project DeepScientist v1.5 is now officially released. It is a local-first open-source autonomous research system for end-to-end scientific discovery. Explore it on GitHub or read the ICLR 2026 paper.
• [2026.03.11] 📄 Our AutoFigure-Edit paper is now available on arXiv and featured in 🤗Hugging Face Daily Papers! If you find our work helpful, please consider giving us an upvote on Hugging Face and citing our paper. Thank you! ❤️
• [2026.02.17] 🚀 The AutoFigure-Edit online platform is now live! It is free for all scholars to use. Try it out at deepscientist.cc or check out our open-source code on GitHub. This new Edit version achieves much better performance!
• [2026.01.26] 🎉 AutoFigure has been accepted to ICLR 2026! You can read the paper on arXiv.

✨ Features

🚀 How It Works

AutoFigure employs a Review-Refine loop to ensure high accuracy and aesthetic quality.

Process:

1. Generate: The agent creates initial SVG/XML based on description & references.
2. Evaluate: The critic scores quality (0-10) and provides specific feedback.
3. Refine: The loop continues until the figure meets publication standards.

🌟 Generated Examples

Here are examples of figures generated by AutoFigure across different domains, showcasing its versatility in handling various levels of complexity.

Category & Visualization
📄 Paper Case
📊 Survey Case
📝 Blog Case
📘 Textbook Case

⚡ Quick Start

Option 1: Python SDK (Recommended)

You can install via cloning the repo:

git clone https://github.com/ResearAI/AutoFigure.git
cd AutoFigure
pip install -e .
playwright install chromium  # Required for rendering

1. Basic Usage (Text-to-Figure)

from autofigure import AutoFigureAgent, Config

# 1. Configure
config = Config(
    generation_api_key="your-api-key",
    generation_provider="openrouter",  # options: 'openrouter', 'gemini', 'bianxie'
    generation_model="google/gemini-3.1-pro-preview",
)

# 2. Generate
agent = AutoFigureAgent(config)
result = agent.generate(
    description="A flowchart showing transformer training pipeline",
    max_iterations=5,
    output_format="svg",
    topic="paper" # 'paper', 'survey', 'blog', 'textbook'
)

print(f"✅ Generated: {result.svg_path} (Score: {result.final_score}/10)")

2. Generate from Paper (PDF/Markdown)

Extract methodology from a paper and generate a figure automatically.

# Generate figure from paper (PDF or Markdown)
result = agent.generate_from_paper(
    paper_path="./paper.pdf",
    max_iterations=5,
    output_format="svg",
    enable_enhancement=True, # Enhance the result
)

if result.success:
    print(f"Extracted methodology: {result.methodology_text[:200]}...")
    print(f"Generated figure: {result.svg_path}")

3. With Image Enhancement

Generate multiple enhanced aesthetic variants of the figure.

result = agent.generate(
    description="Neural network architecture diagram",
    enable_enhancement=True,
    enhancement_count=3,     # Generate 3 variants
    art_style="Modern scientific illustration with clean lines",
    enhancement_input_type="code2prompt" # Best quality mode
)

if result.success:
    print(f"Original Preview: {result.preview_path}")
    print(f"Enhanced variants: {result.enhanced_paths}")

Option 2: Web Interface

Ideally suited for visual interaction and editing.

./start.sh
# Then open http://localhost:6002 in your browser

📊 FigureBench Dataset

We introduce FigureBench, the first large-scale benchmark for generating scientific illustrations from long-form text.

Dataset Overview

Download

from datasets import load_dataset
dataset = load_dataset("WestlakeNLP/FigureBench")

⚙️ Configuration

AutoFigure is highly configurable. You can set these in Config() or via environment variables.

Supported LLM Providers

The Bianxie AI aggregate API provides mainland China-friendly access to AutoFigure's recommended Gemini models without requiring a foreign credit card. Register at bianxieai.

Generation Settings

Methodology Extraction Settings

Enhancement Settings

Pipeline Settings

📚 API Reference

generate() Parameters

generate_from_paper() Parameters

Accepts all parameters from generate() plus:

Result Object (GenerationResult)

Enhancement Modes

📁 Project Structure

AutoFigure/
├── autofigure/              # 📦 Python SDK
│   ├── agent.py             # Main Agent
│   ├── generator.py         # Generation Pipeline
│   ├── enhancer.py          # Image Enhancement
│   └── extractor.py         # PDF Method Extraction
├── frontend/                # 🖥️ Next.js Web UI
├── backend/                 # 🔌 Flask API Server
├── scripts/                 # 🛠️ Utility Scripts
└── pyproject.toml           # Config

🤝 Community & Support

WeChat Discussion Group
Scan the QR code to join our community. If the QR code has expired, please add WeChat ID nauhcutnil or contact tuchuan@mail.hfut.edu.cn. When sending a friend request, please use the format "University/Company - Degree/Position - Name" as your remark.

📜 Citation & License

If you use AutoFigure, AutoFigure-Edit, or FigureBench in your research, please cite:

@inproceedings{
zhu2026autofigure,
title={AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations},
author={Minjun Zhu and Zhen Lin and Yixuan Weng and Panzhong Lu and Qiujie Xie and Yifan Wei and Sifan Liu and Qiyao Sun and Yue Zhang},
booktitle={The Fourteenth International Conference on Learning Representations},
year={2026},
url={https://openreview.net/forum?id=5N3z9JQJKq}
}

@misc{lin2026autofigureeditgeneratingeditablescientific,
      title={AutoFigure-Edit: Generating Editable Scientific Illustration}, 
      author={Zhen Lin and Qiujie Xie and Minjun Zhu and Shichen Li and Qiyao Sun and Enhao Gu and Yiran Ding and Ke Sun and Fang Guo and Panzhong Lu and Zhiyuan Ning and Yixuan Weng and Yue Zhang},
      year={2026},
      eprint={2603.06674},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2603.06674}, 
}

Repository metadata and usage guidance:

• CITATION.cff
• Citation and attribution guidance
• Name and logo usage

🙏 Acknowledgments

We thank the following sponsor for supporting this project:

This project is licensed under the MIT License - see LICENSE for details. Name and logo usage are covered separately in TRADEMARK.md.

More From ResearAI

Explore more open-source research tools from ResearAI:

The optimal configuration for this project uses gemini-3.1-flash-image-preview from Google AI Studio [https://aistudio.google.com/] as the image generation model and gemini-3.1-pro-preview as the Text model. Each run costs approximately $0.50, consumes about 30,000 tokens, and takes around 20 minutes.

[Mainland China Notice] Gemini's Terms of Service do not permit access or usage by users in mainland China. If OpenRouter throws an error, it is often because an account registered in mainland China lacks the necessary permissions to use Gemini. It is recommended to use an OpenRouter account registered in the United States or Europe and to ensure compliant usage.