FYI I'm getting "Too many signups right now. Please try again in a few minutes." when trying to sign up to the waiting list. (congrats haha, but good to fix)
Just what humanity needed: TikTok for scientific papers, with AI! I find myself looking up to the sky wishing for an asteroid to hit Earth on a daily basis, lately...
In some ways I like the concept. Making interesting papers easier to find and easier to digest seems like a good thing.
But the popularity metrics and AI aspects seem like they will cause a bias towards certain types of papers, making potentially useful ones not get found.
Just wanted to maybe make a light suggestions that, for marketing purposes, this really doesn't need any suggestion of TikTok and also might benefit from less heavy handed mentions of AI. I think it provides a real value proposition on its own without needing to rely on those two things to sell itself. They are pretty polarized terms at this point and I can sort of understand the initial revulsion from hearing TikTok next to scientific papers.
I've enjoyed consuming information about interested research papers on instagram, and insta has been good at showing me more of such content. But I think a dedicated platform would be great too! It takes such scientific content creators lots of time to create a script, hook, include animations or other visual aids and also put the research in perspective with it's potential implications in the long terms. I am not sure if AI would be able to do a good job (yet).
My $0.02 try creating an AI powered science channel on YT or insta before spending time on creating a dedicated app.
I like the idea. As others suggested it might be a good idea to drop the branding. Had the same considerations when I built a “Tinder” (1) for RSS Feeds. In the end it worked fine, if not better.
I love this and wanted to build this - but https://www.alphaxiv.org/ already exists, and it gets no social action (hardly any papers have comments), so this makes me doubtful about this.
I am interested to hear if anyone knows why the format may not resonate with researchers or those reading papers in general?
My own reason is that to get value from a "social" site the number of interactions has to be high and of a fast speed for people to continue to engage, which is maybe not possible to hit on research papers.
> My own reason is .. maybe not possible to hit on research papers.
I think fancy people with appropriate credentials and .edu emails are all using openreview? So the audience is what, the unwashed masses who also happen to be doing some light reading at the bleeding edge of knowledge? Surely there are dozens of us I tell you, dozens! =P But yeah, maybe not enough to sustain a social network.
Never heard of alphaxiv, will try. I would also love for this to work, probably not willing to risk slogging through science twitter/bluesky/mastodon. Honestly HN would be the obvious place if it would add a pretty simple tagging system as most of the people interested are probably already here. I don't think we'll see that, because if we had filters no one would go to the front page, and that'd be a bad thing for certain interests.
I follow a few scientists on instagram and it might be a good idea to message them and ask what might be missing. Personally, I like them only because they translate complex ideas into an info that might not be even mentioned in paper - how exactly normal human being without PHD will benefit it. Good example is a recent time crystal news. I would miss it unless one of those cool people on insta would explain. And it might be a good focus for such app.
I'm unsure that the tiktok model works because it's designed around fast, easy to consume content, whereas scientific papers require sitting down and really digesting the material. It's much easier to read dense text on a desktop/tablet over mobile. The times where I read arxiv on mobile, it's really just the abstract. If you summarize each abstract into concise bullet points that might be quite useful.
What make TikTok, well TikTok, is the frictionless experience.
When I opened the link, I expected to directly be shown the target content. If there's a login screen or any explanation to do, it should either be postponed or integrated into the experience.
What I don't understand about these specialized social networks, that obviously won't exist in a few months as they won't get traction, is why not just use the existing social networks?
Instead of some LinkedIn / TikTok / Facebook / Insta for X, create a group or channel in an existing network. Create a subreddit, or Facebook group or telegram channel. There are a number of existing social networks that are good at creating sub-communities. I don't want to join another social media platform.
I think the AI portion is not just something that ought have a toggle, but it should not be part of the platform.
Somewhat recently, the ACM (one of the premier publishers for computer science) integrated AI-generated summaries for all papers, and it made these summaries appear in place of author-written abstracts; to find the abstract, users had to use a toggle. The ACM argued that this was a benefit. After significant community pushback, the ACM has swapped things: author-written abstracts now appear first, but users are still offered a toggle to access AI-generated summaries instead.
As highlighted by professor Anil Madhavapeddy [1], the AI summaries are often factually incorrect, sometimes obviously, but often subtly. This sentiment was corroborated by numerous colleagues of mine less publicly: they checked the AI-generated summaries of their own papers, and for almost every paper were able to identify at least one factually incorrect or significantly misleading statement.
Some people argue that AI-generated summaries help to democratize academia; I think instead they are democratizing misunderstanding. The models fundamentally lack the capacity to "understand" when what they say is wrong or misleading. It is not uncommon that I have students in office hours with severe misgivings about our course material because they asked an LLM some innocuous question to which they thought surely the LLM would generate an accurate response. The course material is, of course, drawn from various sources, so the LLM ought be fairly likely to generate accurate responses. In contrast, a publication is often (or, by definition in my field, necessarily) introducing novel conclusions; this means that the LLM is less likely to generate an accurate summary for a paper than for course materials, and the course material summaries are already problematic enough, so I think applying this to research is just a bad move.
I understand the appeal. I understand how liberating it must feel to someone to get to "talk to" a paper to seek greater understanding. But if you already don't know enough about the material that this is useful, you also don't know enough to know when the responses are subtly incorrect, and I think this completely undermines the purpose of publication in the first place.
55 comments
[ 2.7 ms ] story [ 63.5 ms ] threadFYI I'm getting "Too many signups right now. Please try again in a few minutes." when trying to sign up to the waiting list. (congrats haha, but good to fix)
This looks amazing. I hope Android will be an option.
But the popularity metrics and AI aspects seem like they will cause a bias towards certain types of papers, making potentially useful ones not get found.
Is the gravity set very high or am I getting too old to play Flappy Bird with Transformers?
My $0.02 try creating an AI powered science channel on YT or insta before spending time on creating a dedicated app.
(1) https://philippdubach.com/posts/rss-swipr-find-blogs-like-yo...
I am interested to hear if anyone knows why the format may not resonate with researchers or those reading papers in general?
My own reason is that to get value from a "social" site the number of interactions has to be high and of a fast speed for people to continue to engage, which is maybe not possible to hit on research papers.
I think fancy people with appropriate credentials and .edu emails are all using openreview? So the audience is what, the unwashed masses who also happen to be doing some light reading at the bleeding edge of knowledge? Surely there are dozens of us I tell you, dozens! =P But yeah, maybe not enough to sustain a social network.
Never heard of alphaxiv, will try. I would also love for this to work, probably not willing to risk slogging through science twitter/bluesky/mastodon. Honestly HN would be the obvious place if it would add a pretty simple tagging system as most of the people interested are probably already here. I don't think we'll see that, because if we had filters no one would go to the front page, and that'd be a bad thing for certain interests.
When I opened the link, I expected to directly be shown the target content. If there's a login screen or any explanation to do, it should either be postponed or integrated into the experience.
Instead of some LinkedIn / TikTok / Facebook / Insta for X, create a group or channel in an existing network. Create a subreddit, or Facebook group or telegram channel. There are a number of existing social networks that are good at creating sub-communities. I don't want to join another social media platform.
Somewhat recently, the ACM (one of the premier publishers for computer science) integrated AI-generated summaries for all papers, and it made these summaries appear in place of author-written abstracts; to find the abstract, users had to use a toggle. The ACM argued that this was a benefit. After significant community pushback, the ACM has swapped things: author-written abstracts now appear first, but users are still offered a toggle to access AI-generated summaries instead.
As highlighted by professor Anil Madhavapeddy [1], the AI summaries are often factually incorrect, sometimes obviously, but often subtly. This sentiment was corroborated by numerous colleagues of mine less publicly: they checked the AI-generated summaries of their own papers, and for almost every paper were able to identify at least one factually incorrect or significantly misleading statement.
Some people argue that AI-generated summaries help to democratize academia; I think instead they are democratizing misunderstanding. The models fundamentally lack the capacity to "understand" when what they say is wrong or misleading. It is not uncommon that I have students in office hours with severe misgivings about our course material because they asked an LLM some innocuous question to which they thought surely the LLM would generate an accurate response. The course material is, of course, drawn from various sources, so the LLM ought be fairly likely to generate accurate responses. In contrast, a publication is often (or, by definition in my field, necessarily) introducing novel conclusions; this means that the LLM is less likely to generate an accurate summary for a paper than for course materials, and the course material summaries are already problematic enough, so I think applying this to research is just a bad move.
I understand the appeal. I understand how liberating it must feel to someone to get to "talk to" a paper to seek greater understanding. But if you already don't know enough about the material that this is useful, you also don't know enough to know when the responses are subtly incorrect, and I think this completely undermines the purpose of publication in the first place.
[1] https://anil.recoil.org/notes/acm-ai-recs