Recently, the hundreds of daily updates from arXiv make me feel overwhelmed. To help me better digest new machine learning papers, I developed an arXiv paper reading web app: https://arxiv.gtflashlab.com/ .

It gives me an Outlook-like user experience. It allows me to view the latest papers in an efficient, manageable, personalized way. It also provides some basic functions, that help me to manage these papers:

  • Reading status: it tells me whether I have viewed the paper or not.
  • Archive: I can archive papers that I don’t want to see anymore.
  • Star: Save my favorite papers.
  • Tagging: it allows me to add tags to categorize the papers.
  • Tagging from abstract/title: I can efficiently add tags from the title/abstract by simply selecting the text.
  • Filtering by Reading status/Archive status/Star/Tags
  • Read Abstract/PDF/HTML
  • PDF Annotation which can be saved on the server.
  • Search: Full-text-search on title and abstract (If you can’t find the paper, add “&&searchEngine=simple” to the URL. It will perform a case-insensitive exact-match search, which is kind of slow. Don’t blame me for the search experience, I am a newbie to the database. ) Citation/Reference (based on semantic scholar)
  • View papers on your iPhone/iPad by saving the website as a bookmark on the homepage. (Although some things are still malfunctioning. /(ㄒoㄒ)/~~)

I want to share it with anyone who might find it useful for doing research.

Acknowledge It is powered by arXiv API, semantic scholar API, arXiv Vanity, and Adobe PDF API. This project is still in an early stage and I only import the machine learning related papers. If you want to help me to improve it, visit my GitHub: https://github.com/HMJiangGatech/ArxivRoller