yu3zhou4

16 ms·

1. 211.▲
   by yu3zhou4 2y ago

   Jakub Pachocki is still in OpenAI though
2. 212.▲
   by yu3zhou4 2y ago

   It would be great to get an explanation for each of items here for non-C++ devs
3. 213.▲
   by yu3zhou4 2y ago

   Thank you, dang!
4. 214.▲
   by yu3zhou4 2y ago

   That would make sense. I think how to approach it, so we can have collectively built definitions of concept like NeRF. Maybe we can treat them in a similar way to TL;DRs - everyone can write their own version and we can vote on the best one
5. 215.▲
   by yu3zhou4 2y ago

   This is a neat idea, I never thought about filling gaps with hyperlinks before. I will explore this idea further, thank you so much rsfern!
6. 216.▲
   by yu3zhou4 2y ago

   Core points to communicate after gaining insights from your comments: - platform is open to abstracts (TL;DRs) for all subjects of research - AI-generated TL;DRS will be soon integrated to the platform (initially as a fallback when there is
7. 217.▲
   by yu3zhou4 2y ago

   Can it be still added? @dang?
8. 218.▲
   by yu3zhou4 2y ago

   A brand new Discord (it's so new that you can be a #2 member :D) https://discord.gg/AJ9YCcqD It seem to be a go-to choice platform for communities, so let's try it. You can request features and changes and I post
9. 219.▲
   by yu3zhou4 2y ago

   Here's how I picture this, in an optimistic scenario: Once the number of abstracts grows up, we vote up and down on them, so at the end you get the best abstract for each paper, which is supposed to be informative enough to give you a
10. 220.▲
    by yu3zhou4 2y ago

    The holy grail would be generating TL;DRs with a model trained on the best human-written abstracts. Although - there should be a way to hide machine-generated content, for your convenience, since not everyone might be interested in this con
11. 221.▲
    by yu3zhou4 2y ago

    Yes! After finishing the most urgent features like sorting, filtering and fixing some bugs, I'd like to make it easy to subscribe to a chosen subject to get a relevant papers to your mailbox
12. 222.▲
    by yu3zhou4 2y ago

    Hi, sorry for the confusion It started as a TL;DRs platform for research papers on AI, ideated by Francois Fleuret on X, but when I started coding this (and of course already bought a domain :p), I realized that there's no reason to li
13. 223.▲
    by yu3zhou4 2y ago

    Laughed hard on that :D Sorry if the wording is not great, I'm neither a native English speaker nor good at copy-writing. I worry the title can't be changed anymore anyway
14. 224.▲
    by yu3zhou4 2y ago

    Thanks! Just a basic stuff, Next.js on Vercel, Auth0, Postgres and great UI components by shadcn
15. 225.▲
    by yu3zhou4 2y ago

    Hello, much thanks for that. AI-generated abstracts might help with filling the platform with a content and as you mentioned, it might be less biased and more objective. This is something people on X also discussed below the launch post [0]
16. 226.▲
    by yu3zhou4 2y ago

    Hello! You are on point, building community to write the abstracts, while also setting up relevant guidelines are my main focus for now. I will think of how to communicate guidelines and expectations to the content in a clear way. Thank you
17. 227.▲
    Show HN: Community-written abstracts for research papers(tldr-ai.org)

    79 points by yu3zhou4 2y ago|51 comments
18. 228.▲
    by yu3zhou4 2y ago

    I got a hunch that a free will is related to both self-control and self-awareness
19. 229.▲
    by yu3zhou4 3y ago

    Hey Sławomir, great work! Do you think it can be somehow adjusted to the youngest kids that are not ready for Python, but would like to code anyway? Powodzenia, trzymam kciuki za Ciebie i Twoją platformę!
20. 230.▲
    Show HN: ai("question", data) infers the ML model and answers with Python type(github.com)

    1 points by yu3zhou4 3y ago|0 comments
21. 231.▲
    by yu3zhou4 3y ago

    Here's the repo: https://github.com/SuperMayo/mayerowitz.io
22. 232.▲
    by yu3zhou4 3y ago

    It's like a combination of HuggingGPT, LangChain and Python type inference
23. 233.▲
    Case study of building an AutoML library that operates natural language(maczan.pl)

    2 points by yu3zhou4 3y ago|1 comments
24. 234.▲
    Show HN: Text-to-ML: like HuggingGPT + LangChain + type inference(github.com)

    1 points by yu3zhou4 3y ago|0 comments
25. 235.▲
    Show HN: I just open sourced my CSV-to-ML AutoML App(github.com)

    1 points by yu3zhou4 3y ago|0 comments
26. 236.▲
    by yu3zhou4 3y ago

    A quite interesting read. It gave me Prolog vibes.
27. 237.▲
    by yu3zhou4 3y ago

    A short but nice experience, with pleasant UI and a bit of fun of prompting. Thanks for sharing and good job :) Maybe try longer forms as well because maybe you might find yourself in writing scifi
28. 238.▲
    by yu3zhou4 3y ago

    The platform infers about columns you provided and picks the one that is most likely the one you’d like to predict about
29. 239.▲
    by yu3zhou4 3y ago

    Thanks a lot! Hopefully I’ll be able to share more details in form of a blog post or short paper later this year I’d like to not share actual details yet, since this platform is a starting point of a broader research and development I’m doi
30. 240.▲
    by yu3zhou4 3y ago

    Thank you so much for the comment, I think you’re on point and I’ll work on a better end-to-end demo on YouTube. Perhaps it will be scripted, but certainly more informative :)