AutoResearcher is an open-source project that uses GPT-based AI models to automatically generate academic literature reviews based on a given research question. The script fetches top papers from the Semantic Scholar API, extracts relevant information, and combines the findings into a concise literature review.
The project is a very early prototype and is still under development. The vision is to create a tool that can conduct actual scientific discovery on autopilot.
- Clone the repository:
git clone https://github.com/eimenhmdt/autoresearcher.git
- Create a virtual environment and activate it:
cd autoresearcher
python3 -m venv venv
source venv/bin/activate
On Windows, use venv\Scripts\activate
instead of source venv/bin/activate
.
- Install the required Python packages:
pip install -r requirements.txt
- Create a .env file in the project directory and add your OpenAI API key and an email of your choice (used to identify your API requests for getting citations):
OPENAI_API_KEY=<your_openai_api_key>
EMAIL=<your_email>
Replace <your_openai_api_key> with your actual API key from OpenAI.
- Open the main.py file and set your research question at the bottom of the script:
research_question = "<your_research_question>"
Replace <your_research_question> with your desired research question.
- Run the script with your research question as a command-line argument:
python auto_researcher.py "<your_research_question>"
Replace <your_research_question> with your desired research question. If no argument is provided, the script will use the default research question.
The script will fetch the top papers, extract answers, and generate a literature review.
Contributions are welcome! Please feel free to submit issues or create pull requests. Let's take upgrade science together! ๐
This project is licensed under the MIT License. See the LICENSE file for details.
Made with coffee by @eimenhamedat