Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: AskHN

patterns.app

Post reply on HN
Print view View original

121–130 of 139 posts

Re: Show HN: AskHN

#121

Post by setgree »

soheil wrote: post #114
setgree wrote: post #88

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]

Her essay is from 2019, and I quoted it mainly to say that I am not making an original point, this has been a known problem with LLMs for a while (and I presume it will continue to be).

Re: Show HN: AskHN

#122

Post by kerogerokero »

Anyone here know how to generate subtitles automatically using AI when a video is playing on the web?

Was planning to see how I can build something like these sites but without the need to regularly update the subtitles catalog: https://subscene.be https://subtitlecat.com https://subtitletrans.com https://my-subs.co

Re: Show HN: AskHN

#123

Post by hummus_bae »

kerogerokero wrote: post #122

Anyone here know how to generate subtitles automatically using AI when a video is playing on the web? Was planning to see how I can build something like these sites but without the need to regularly update the subtitles catalog: https://subscene.be https://subtitlecat.com https://subtitletrans.com https://my-subs.co

[dead]

Re: Show HN: AskHN

#124

Post by gnulinux »

soheil wrote: post #114
setgree wrote: post #88

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]

A simple Google search yields that she's likely an expert and her opinion can be used as a "citation" in one's comment:

> Brief bio: I started out studying math (Princeton AB ‘10, Yale PhD ‘15, focusing on applied harmonic analysis) and then spent some time in the world of data science and machine learning (Palantir, Recursion Pharmaceuticals, Starsky Robotics.)

from: https://srconstantin.github.io/about/

Re: Show HN: AskHN

#125

Post by seanhunter »

soheil wrote: post #114
setgree wrote: post #88

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]

Giving credit to someone for the quotation is called a citation, and it's encouraged.

Re: Show HN: AskHN

#126

Post by underlines »

Just to be sure: This is NOT a finetuned GTP model, but rather standard GPT-3 API, used to summarize search results of a HN Comments DB, based on user input. Right?

Re: Show HN: AskHN

#127

Post by 2020aj »

setgree wrote: post #98
leereeves wrote: post #95

Earlier quoted context omitted.

> there's some nonsense interspersed here, e.g. the claim that R has "a more mature package universe" than Python. As a programmer, I find R hard to use and not very well designed, so I can see why you'd call that nonsense. But when I was a math student, I found that in some ways R does have "a more mature package universe". There were many math algorithms that I could find packages for in R and not in Python, even a…

Absolutely, for statistics and visualization I think R and its packages are (sometimes) superior. But GPT responses don't generally offer those kinds of nuances; the claim is that the packages are "more mature," period. And it's for good reason that the _most_ mature Python packages, e.g. numpy and pandas, are used by data scientists in production pretty much everywhere.

Regarding numpy/pandas: What's the reason outside of them being _in Python_?

Re: Show HN: AskHN

#128

Post by rocho »

bityard wrote: post #14
retube wrote: post #3

Earlier quoted context omitted.

I am guessing they already were? But this is 100% pure, concentrated HN not contaminated with nonsense from the rest of the web :)

If it's really trained exclusively off of HN comments, I expect most of the bot's responses will evade the actual question but spend several paragraphs debating the factual specifics of every possible related tangential point, followed by an thinly-veiled insult questioning the user's true motivations.

It's not trained at all. The bot finds relevant comments and then uses OpenAI's API to summarize them.

Re: Show HN: AskHN

#129

Post by _8j50 »

Some day I will sue people like OP (if they're monetizing it) and OpenAI for monetizing my public posts. You can use, reuse and alter public speech but when you earn ad dollars...yeah part of that is mine if your model used my public content. I probably won't actually sue but someone will.

I am not a lawyer but there has to be a jurisdiction where I can establish standing at least.

Re: Show HN: AskHN

#130

Post by nine_k »

shanusmagnus wrote: post #108
setgree wrote: post #88

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..…

I wonder if the "default to humility" heuristic probably does more harm than good on net, because the people who heed it probably shouldn't, and the ones who should won't.

Default to humility. Do not assume you're so smart that you can skim the text and understand it correctly. Read every word, don't assume that the author is so predictable that you can guess correctly.

Why, does not sound too arrogant to me.

Post reply on HN