The app works for 1) arXiv papers 2) any url of a paper obtainable from the address bar of your phone.
If you copy-paste, or copy and use the clipboard button, it will try to identify the arXiv identifier from the url, and then show you the title + abstract. Then you have to download the PDF, preview the transformed copy (copy stays in the cloud for 24h), and finally request a fully reflowed copy. The reflow is limited to about 30s via AWS Lambda, so this is not for dissertations -- this is for 2-20 page papers you can reasonably read on your phone. Nevertheless, there is a progress bar at the top of the screen to show you it's working, at that stage. These steps are pretty much reproduced in the app store images.
For a non-arXiv link, the only difference is that there is no abstract. But note, the link can't just be to the journal page, you've got to get a PDF mime-type when it's requested.
There are some things I should change, but it's very useful to me, and I just verified it works for both arXiv and non-arXiv. If it's not working for you, I would suggest deleting and reinstalling (sorry!).
edit: fyi it's originally because the information about Encryption Compliance is so confusing. I just use HTTPS and I'm pretty sure the documentation has gotten clearer in the meantime, so I'm optimistic.
I wrapped this for iOS, and linked it to Twitter. If you connect it to your account, there is a separate feed of any tweet with a PDF, and one for any arXiv or PDF tweet that you've liked. [You don't have to connect, you can also just input the URL or quick-paste].
It's great for following your favorite scientists and professionals, and now I favorite more of their posts, which makes me happy because I feel like I am spreading more love in the world.
Interesting that the draft was predetermined (unlike the previous exhibition, which was live), allowing OpenAI to train ahead of time on a restricted set of matches. I wonder how long that was?
If you copy-paste, or copy and use the clipboard button, it will try to identify the arXiv identifier from the url, and then show you the title + abstract. Then you have to download the PDF, preview the transformed copy (copy stays in the cloud for 24h), and finally request a fully reflowed copy. The reflow is limited to about 30s via AWS Lambda, so this is not for dissertations -- this is for 2-20 page papers you can reasonably read on your phone. Nevertheless, there is a progress bar at the top of the screen to show you it's working, at that stage. These steps are pretty much reproduced in the app store images.
For a non-arXiv link, the only difference is that there is no abstract. But note, the link can't just be to the journal page, you've got to get a PDF mime-type when it's requested.
There are some things I should change, but it's very useful to me, and I just verified it works for both arXiv and non-arXiv. If it's not working for you, I would suggest deleting and reinstalling (sorry!).