This tool converts PDF research papers into XMind mind maps, leveraging ChatGPT to summarize content, extract images, and render formulas, thereby enhancing paper reading efficiency for researchers and students.

How It Works

The system processes PDF documents, using regular expressions to identify and extract key sections like abstracts and introductions, along with figures and equations. It then utilizes ChatGPT (or a compatible LLM) to generate concise summaries and hierarchical outlines. Image and equation extraction can be handled by PDFFigure2 (requiring a Java environment) or alternative methods, with options to include captions. The output is structured as an XMind file.

Quick Start & Requirements

• Install: Clone the repository (git clone --recursive), then install dependencies using pip install -r requirements.txt and pip install -r ./XmindCopilot/requirements.txt.
• Prerequisites: Python 3.x, OpenAI API key, and optionally Java Development Kit (JDK) if USE_PDFFIGURE2 is enabled for image/table extraction.
• Usage: Run python paper2xmind.py --path <pdf_path_or_folder>.
• Configuration: Modify config.py for API keys, model selection, proxy settings, and generation options (images, equations, captions).
• Docs: Configuration details are in config.py.

Highlighted Details

• Supports custom OpenAI API endpoints and multiple API keys for parallel processing.
• Configurable to extract images, equations, and table captions.
• Allows disabling ChatGPT for summary generation, focusing solely on structural elements.
• Includes options for proxy configuration for API access.

Maintenance & Community

The project is maintained by Master Yip. A QQ group is available for community discussion.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive for commercial use and integration with closed-source projects.

Limitations & Caveats

The accuracy of the generated XMind notes is dependent on ChatGPT's output quality and is best used as a draft. The TEXT2LIST_MAX_NUM and TEXT2TREE_MAX_NUM parameters are noted as temporarily ineffective. Proper setup of XmindCopilot is crucial, as indicated by common import errors.