AutoResearch: A General Framework for Academic Research Automation

AutoResearch is a framework designed to simplify and accelerate academic research tasks. It offers a modular and extensible architecture to help researchers, developers, and academics efficiently search, organize, summarize, and analyze academic papers.

See this Google Colab notebook for an installation-free quick demo.

Key Features

  • Automated Paper Search - Search for academic papers using keywords and retrieve metadata from Google Scholar, Semantic Scholar, and arXiv. Organize results by relevance or date, apply filters, and save articles to a specified folder.

  • Paper Summarization - Summarize individual papers or all papers in a folder. Extract key sections (abstract, introduction, discussion, conclusion) and generate summaries using GPT models. Track and display the total cost of summarization.

  • Explain a Paper with LLMs - Interactively explain concepts, methodologies, or results from a selected paper using LLMs. Supports user queries and detailed explanations of specific sections.

  • Code Availability Check - Check for GitHub links in papers and validate their availability.

  • Topic-to-Survey Automation - Convert a topic or research question into a comprehensive survey of relevant papers. Generate keywords, retrieve articles, summarize content, and optionally check code availability. Organize and zip results for easy access.

Basic Example

from auto_research.search.core import AutoSearch
from auto_research.survey.core import AutoSurvey

# Search for papers
search = AutoSearch(keywords="machine learning", num_results=10)
search.run()

# Summarize a paper
survey = AutoSurvey(api_key="your-api-key", model="gpt-4", paper_path="path/to/paper.pdf")
survey.run()

See Examples for more advanced usages

Communication

Contribution

We welcome contributions! Please see the Contribution Guide for details.

License

MIT License. See LICENSE for details.

Contents:

Indices