It doesn't have enough ratings yet to judge what users think.
PaperFeed - arXiv Paper Reader is made and published by
勇 刘.
AppBrain is an independent app data site and isn't affiliated with 勇 刘.
The app name, icon, screenshots and description belong to 勇 刘 and are shown as
published on the App Store.
About PaperFeed - arXiv Paper Reader
PaperFeed - arXiv Paper Reader is an education app developed by 勇 刘.
User Ratings
PaperFeed - arXiv Paper Reader is rated 5.00 out of 5 stars. The rating is based on one rating ratings.
App Information
PaperFeed - arXiv Paper Reader is freely available from the App Store. The app size is 18.22 MB. The latest version available is 1.7.0. The last update was on October 11, 2026.
Technical Requirements
The app has a content rating of Low Maturity. The app has been available on the App Store since March 2026.
Description by 勇 刘(from the App Store)
How long does it take to read one English paper? Skim the abstract, guess at the method, then force yourself through the PDF anyway. PaperFeed compresses that into three steps: let an AI summary tell you whether a paper is worth your time, read it paragraph by paragraph in two languages, and ask the paper itself wherever you get stuck — every answer cites the source, so you can verify before you trust.
**Discover: from "what's new" to "what's worth reading"** • Follow the arXiv categories you care about, then switch between four feeds: latest papers, highly cited classics, community favorites, and recommendations picked for you • A daily pick chosen for you, delivered at 9 AM in your time zone — tap to jump straight into the paper • Search by keyword, time range, and category, with suggestions and search history
**Understand: more than translation** • Bilingual reading, paragraph by paragraph: the translation sits between the original paragraphs, with formulas, figures, and terms in place • Structured AI summaries: research problem, method, key results, and limitations — one paper in one minute • Chat with the paper: full-text retrieval-augmented answers with source citations and confidence levels, multi-turn follow-ups included • Deep dive: have AI walk you through a paper's method and contribution in plain language • Audio overviews: a narrated walkthrough of a paper's problem and approach — variable speed, line-by-line captions, a push when it's ready (1 free per day, unlimited on Pro) • AI explainer videos: a narrated video that walks through the method and contribution in your language — generation takes 10–20 minutes, with a push notification when it's ready (Pro) • Highlights, table of contents, font size and themes, and automatic reading progress
**Everything around the paper** • Citation and reference networks: where a paper comes from and where it leads • Code ecosystem: linked GitHub repos and Hugging Face models and datasets • Conference reviews: public OpenReview reviews from venues like ICLR and NeurIPS (where available)
**Who it's for** Researchers, graduate students, engineers, and anyone who needs to read papers efficiently. You don't have to finish the PDF to know what a paper is worth — let the parts that matter surface first.
**Subscription** • Free: 3 AI uses per day (+1 bonus when you share), 1 audio overview per day, bilingual preview of every paper • Pro: 600 AI uses per month, full bilingual papers, unlimited audio overviews, AI explainer videos, all AI models unlocked • The annual plan saves more. Prices as shown in the app.
Interface available in English, Chinese (Simplified & Traditional), Japanese, Korean, German, Spanish, French, and Portuguese (Brazil). Translations support 30+ languages.
Questions or feedback? Visit https://paper.toai.pro or contact support@tanbox.site
Terms of Use (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
Each subscription will automatically renew 3 days before the expiration date for
the same time period. Subscriptions can be cancelled at any time before the renewal.