111–120 of 139 posts
Nice! We built something very similar recently, it is more like "Ask your documentation" but very similar implementations otherwise. See a demo on the huggingface transformers documentation: https://huggingface.co/spaces/jerpint/buster code: https://github.com/jerpint/buster
Am I correct in understanding that this doesn't actually generate answers based on HN, but instead finds semantically-near comments and sends them verbatim to GPT to summarize? Seems like a good enough hack, though I'd love a detailed writeup of how to actually specialize an existing LLM with additional training data (like HN).
You can literally finetune these OpenAI models using their API. In this case it probably wasn't done because the author found it too much work and/or too expensive.
Military budgets and manpower would make sock puppets interesting. https://www.theguardian.com/technology/2011/mar/17/us-spy-op...
Does anyone find it curious Edward Snowden is mute on the military's use of this technology?
How many people would be surprised to learn they are communicating with a program on websites like this?
As is often true of GPT responses, there's some nonsense interspersed here, e.g. the claim that R has "a more mature package universe" than Python. I think this is false, but if you're reading quickly, it sounds cogent enough. As Sarah Constantin observed about GPT2 [0]: > if you skim text, you miss obvious absurdities. The point is OpenAI HAS achieved the ability to pass the Turing test against humans on autopilot..…
As is often true of GPT responses, there's some nonsense interspersed here, e.g. the claim that R has "a more mature package universe" than Python. I think this is false, but if you're reading quickly, it sounds cogent enough. As Sarah Constantin observed about GPT2 [0]: > if you skim text, you miss obvious absurdities. The point is OpenAI HAS achieved the ability to pass the Turing test against humans on autopilot..…
Earlier quoted context omitted.
Nah, it's no big a deal, its not like cambridge analytica will happen again. They're just using your data to train AI. Who knows may be based on the way you comment, you may get suggestions on which medication you need, or if it's time for the Redbull/starbucks coffee. Nah, all is good. Nothing bad will happen in allowing companies to scrape comments and build models. They're very ethical. In fact, people here are su…
The BIG DEAL is not THAT specific instance but the fact that the ML crowd think it's OK to take everything without even asking permission
Everything they take was freely given. Thrown into the void. Screamed into the wind. It's weird that people are perfectly fine if someone happens to read their words (at all) and fine if some of those who do read them manage to find something in them that is in any way helpful or useful, but the moment they think someone else might make money as a result of something gained from exposure to those same words it's somehow offensive and everyone starts demanding a cut of (usually non-existent) profit.
The "ML" crowd has just as much a right to read and learn from the words I enter on social media platforms as anyone else. I'm not charging any kind of fee for the words of debatable wisdom, fact checking, or shitposting I "contribute". I didn't ask permission before replying to your comment. Why should anyone feel like they should ask for permission from me to read it? What exactly is "taken" from me beyond the time I voluntarily spent participating in online discourse?
Let's admit that HN's culture is that many of us are confidentially wrong, which we cover up with impressive technical jargon. As such, any wrong answer in this AI is in fact correct.
I get the feeling this comment is self-referential/self-parodying.
I'm quite pleasantly surprised by this response: Question: what are the biggest fallacies of people that post on hackernews? Answer: One of the biggest fallacies of people that post on Hacker News is the nirvana fallacy, which is when someone jumps from "thing x is imperfect" to "thing x is worthless" without considering any of the benefits. People also tend to be familiar with certain technologies and will discount…
Is that really a problem? It's only natural to get accustomed to old problems caused by old tech (where workarounds probably already exist or limitations understood and accepted) while it makes sense to hold new tech to a higher standard since it has the benefit of learning from past mistakes and should avoid introducing obvious problems for people in the future to deal with. Our future selves will thank us for being more critical of flawed new tech than we are of old tech.
Related question: I've written probably a million words over my lifetime. Is there an easy way to load up GPT with my thoughts to have it be a fake me?
As is often true of GPT responses, there's some nonsense interspersed here, e.g. the claim that R has "a more mature package universe" than Python. I think this is false, but if you're reading quickly, it sounds cogent enough. As Sarah Constantin observed about GPT2 [0]: > if you skim text, you miss obvious absurdities. The point is OpenAI HAS achieved the ability to pass the Turing test against humans on autopilot..…
[flagged]