Showcase of Collaborative Projects at arXiv Labs

  1. arXiv Bibliographic Findr
    • Tech Insight: Developed in collaboration with Matt Bierbaum from Cornell Computing and Information Science, this tool offers a dynamic way to explore the citation network of arXiv papers. It’s designed for seamless navigation through an article’s citation tree, enhancing the discovery of relevant research.
  2. ar5iv
    • Tech Specs: A collaboration with Michael Kohlhase and Deyan Ginev from Friedrich-Alexander Universität, Bruce Miller from NIST, and Ben Firschman from arXiv-Vanity. This project focuses on enhancing the usability and accessibility of arXiv papers by converting them into HTML format, a step towards improved readability and wider accessibility.
  3. CORE Recommender
    • Behind the Scenes: Working with the CORE Team, this feature recommends relevant open access papers from a global network, including over 10,000 data providers. It’s a sophisticated aggregation tool that connects users with a vast array of research papers, right from the arXiv interface.
  4. arXiv Links to Code & Data
    • Tech Details: In partnership with Robert Stojnic, Viktor Kerkez, and Ludovic Viaud from Papers with Code/Meta AI Research, this project provides a streamlined way to find relevant code for a paper. It leverages the extensive database of Papers with Code, linking papers to code and results in Machine Learning.
  5. Connected Papers
    • Tech Framework: Developed by Alex Eitan Tarnavsky and team from Connected Papers, this unique tool offers a visual method to explore papers relevant to a specific field. It’s an innovative way to map out academic landscapes, create bibliographies, or just navigate the world of academic papers.
  6. Litmaps
    • Tech Angle: Collaborating with Kyle Webster and the Litmaps team, this tool integrates interactive citation maps and modern search capabilities to offer a comprehensive research discovery experience. It allows users to visualize and navigate the citation network surrounding an arXiv article.
  7. Hugging Face Spaces
    • Tech Implementation: This collaboration with Abubakar Abid and the Hugging Face team brings interactive applications and model demos directly into the arXiv framework. Leveraging open-source tools like Gradio and Streamlit, it allows users to experiment with models and datasets in an accessible, code-free environment.
  8. IArxiv
    • Technical Process: Developed by Ezequiel Alvarez and the Easytech team, IArxiv uses an unsupervised Machine Learning algorithm, LDA, to tailor paper recommendations according to user preferences. It evolves to fit the reader’s interests, providing a personalized arXiv experience.
  9. Scite Smart
    • Technical Insight: A collaboration with Josh Nicholson and Domenic Rosati, this platform enhances the process of discovering and evaluating scientific articles. It introduces Smart Citations, providing context and classification for each citation within a publication.
  10. ScienceCast
    • Tech Overview: In partnership with Andrew Jiranek and ScienceCast, Inc., this project connects arXiv papers to interactive video and audio presentations, including AI-generated summaries, broadening the dissemination of scientific findings.
  11. Replicate
    • Tech Dynamics: Working with Ben Firshman and team, Replicate streamlines the process of associating machine learning models with arXiv papers, making these models more accessible and user-friendly.
  12. Influence Flower
    • Technical Architecture: Developed by Minjeong Shin and Lexing Xie, this tool visualizes citation influences among academic entities, offering a unique perspective on the impact and connections of research works.
  13. DagsHub
    • Tech Focus: A collaboration with Dean Pleban and the DagsHub Team, this platform serves as a central hub for hosting, discovering, and collaborating on research projects, emphasizing full and reproducible implementations of arXiv