# Overview Keeping on top of the latest most interesting scientific publications can be hard. With so many papers and preprints coming out every week, how much time do you want to spend picking the most relevant ones? Hopefully *not* a lot, so you'd rather automate this boring task. Well... that's what **refy** is for :sunglasses: ! {% hint style="info" %} **Refy** downloads the latest preprints from [biorxiv](https://www.biorxiv.org/) and [arxiv](https://arxiv.org/) and checks them against a library of papers you've previously read to select the most relevant preprints for you. {% endhint %} There's other literature recommendation systems out there that aim to facilitate keeping up with the literature (e.g., [meta.org](https://www.meta.org/), [inciteful](https://inciteful.xyz/), [scite.ai](https://scite.ai/)), but these generally work using *keywords* to select papers of interest. This usually is not specific enough and you end up sifting through lots of unrelated papers to find an interesting one. Other softwares use one *single* reference paper instead and look for other papers similar to that. That's **better**, but it's not **great**... Your library likely has hundreds of papers that you've read and, hopefully, found interesting. That's **a lot of information** about what kind of papers you'd want to read next. So why not use this information to select recently published papers that you'd actually want to readd? That's **exactly** what **refy** aims to do. Refy uses a `.bib` file with metadata about *lots* of papers you've read as input. It then uses natural language processing algorithms to find preprints whose abstract is **semantically similar** to those of papers you've read before, at it works :tada: ! **Refy**'s very easy to install and use (see [User guide](/refy/user-guide.md)) and you can even [automate](/refy/user-guide/setting-up-github-actions.md) your literature updates even more. Have fun and get in touch with questions, bugs or suggestions on [GitHub](https://github.com/FedeClaudi/refy). --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://federicoclaudi.gitbook.io/refy/master.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.