| --- |
| license: cc-by-4.0 |
| task_categories: |
| - text-to-image |
| configs: |
| - config_name: default |
| data_files: |
| - split: dev |
| path: data/dev.parquet |
| - split: test |
| path: data/test.parquet |
| --- |
| |
| # FigureBench |
|
|
| **The first large-scale benchmark for generating scientific illustrations from long-form scientific texts.** |
|
|
| [Paper](https://huggingface.co/papers/2602.03828) | [Code](https://github.com/ResearAI/AutoFigure) |
|
|
| ## Overview |
|
|
| FigureBench is curated to encompass a wide array of document types, including research papers, surveys, technical blogs, and textbooks, establishing a challenging and diverse testbed to spur research in automatic scientific illustration generation. |
|
|
| This dataset contains: |
| - **Development Set (dev)**: 3,000 samples with conversation-format annotations for training and development |
| - **Test Set (test)**: 300 samples with simple text-image pairs **strictly for evaluation** |
|
|
| ## HuggingFace |
|
|
| The dataset is available on HuggingFace: [WestlakeNLP/FigureBench](https://huggingface.co/datasets/WestlakeNLP/FigureBench) |
|
|
| ```python |
| from datasets import load_dataset |
| |
| # Load the dataset |
| dataset = load_dataset("WestlakeNLP/FigureBench") |
| |
| # Access development set (for training/development) |
| dev_sample = dataset["dev"][0] |
| |
| # Access test set (for evaluation only) |
| test_sample = dataset["test"][0] |
| ``` |
|
|
| ## Dataset Statistics |
|
|
| | Category | Dev Samples | Test Samples | Avg. Text Tokens | Text Density (%) | |
| |----------|-------------|--------------|------------------|------------------| |
| | Paper | 2,900 | 200 | 12,732 | 42.1 | |
| | Blog | 20 | 20 | 4,047 | 46.0 | |
| | Survey | 40 | 40 | 2,179 | 43.8 | |
| | Textbook | 40 | 40 | 352 | 25.0 | |
| | **Total** | **3,000** | **300** | - | - | |
|
|
| ## Key Challenges |
|
|
| - **Long-context reasoning**: Text tokens vary by over an order of magnitude (352 for Textbooks to 12,732 for Papers) |
| - **High text density**: Average 41.2% of image area occupied by text |
| - **Structural complexity**: Average 5.3 components and 6.4 shapes per illustration |
| - **Visual richness**: Average 6.2 colors per illustration |
|
|
| ## Data Format |
|
|
| ### Development Set (dev) |
|
|
| Each sample in the development set uses a conversation format: |
|
|
| ```json |
| { |
| "messages": [ |
| { |
| "role": "system", |
| "content": "You are an expert in analyzing scientific papers..." |
| }, |
| { |
| "role": "user", |
| "content": "Please determine whether the image is the most important illustration... |
| |
| [Paper text content]" |
| }, |
| { |
| "role": "assistant", |
| "content": "yes" |
| } |
| ], |
| "images": ["images/{paper_id}/{image_filename}"] |
| } |
| ``` |
|
|
| ### Test Set (test) |
|
|
| The test set uses a simple text-image pair format for evaluation: |
|
|
| ```json |
| { |
| "text": "Original document text content...", |
| "image": "test_images/{category}/{filename}.png", |
| "category": "paper|survey|blog|textbook" |
| } |
| ``` |
|
|
| | Field | Type | Description | |
| |-------|------|-------------| |
| | text | string | Original document text (LaTeX for papers, Markdown for blogs/surveys, plain text for textbooks) | |
| | image | string | Relative path to the figure image | |
| | category | string | Document category: paper, survey, blog, or textbook | |
|
|
| ## Usage |
|
|
| ### Loading Development Set |
|
|
| ```python |
| from datasets import load_dataset |
| |
| dataset = load_dataset("WestlakeNLP/FigureBench") |
| dev = dataset["dev"] |
| |
| sample = dev[0] |
| messages = sample["messages"] |
| image_path = sample["images"][0] |
| ``` |
|
|
| ### Loading Test Set for Evaluation |
|
|
| ```python |
| from datasets import load_dataset |
| |
| dataset = load_dataset("WestlakeNLP/FigureBench") |
| test = dataset["test"] |
| |
| sample = test[0] |
| text = sample["text"] |
| image_path = sample["image"] |
| category = sample["category"] |
| ``` |
|
|
| ## Dataset Structure |
|
|
| ``` |
| FigureBench/ |
| ├── README.md |
| ├── data/ |
| │ ├── dev.parquet # Development set (3,000 samples) |
| │ └── test.parquet # Test set (300 samples) |
| ├── images/ # Development set images |
| │ ├── {paper_id}/ |
| │ │ └── {paper_id}_{figure_name}.png |
| │ └── ... |
| └── test_images/ # Test set images |
| ├── blog/ |
| ├── paper/ |
| ├── survey/ |
| └── textbook/ |
| ``` |
|
|
| ## Data Curation |
|
|
| ### Source Data |
|
|
| The dataset is curated from multiple high-quality sources: |
|
|
| - **Papers**: Sampled from the Research-14K dataset, filtered using GPT to select illustrations that best represent core methodologies |
| - **Surveys**: Structural diagrams (roadmaps, taxonomies) from recent AI surveys on arXiv |
| - **Textbooks**: Sourced from open-licensed educational platforms (e.g., OpenStax) for pedagogical clarity |
| - **Blogs**: Hand-collected from technical outlets (e.g., ICLR Blog Track) for modern visual styles |
|
|
| ### Annotation Process |
|
|
| 1. Initial filtering to retain only conceptual illustrations (excluding data-driven charts) |
| 2. Ensured each key visual element is explicitly described in the source text |
| 3. Two independent annotators evaluated each pair |
| 4. Only pairs approved by both annotators were included |
| 5. Achieved high Inter-Rater Reliability (IRR) of 0.91 |
|
|
| ## Citation |
|
|
| If you use FigureBench or AutoFigure in your research, please cite: |
|
|
| ```bibtex |
| @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} |
| } |
| ``` |
|
|
| ## License |
|
|
| This dataset is released under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). |
|
|
| The curation process adhered to open-source licenses. Please refer to the original papers for the copyright of individual figures. |