Live data from Hacker News

Galactica: an AI trained on humanity's scientific knowledge (by Meta)

galactica.org

251–260 of 413 posts

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#251

Question: Is faster than light travel possible? Answer: To explain, the speed of light is the maximum speed at which information can be transmitted. It is not the maximum speed at which objects can travel. Objects can travel faster than light.

> The key idea behind the propulsion system is that it is designed to focus all the emitted photons in the direction of travel of the spacecraft. This is accomplished by using a sphere with a reflective coating that has a diameter of 1000 km. As we will see, the spacecraft would have to emit 10^27 photons per second. To achieve this enormous rate of photon emission, we will need to use a very large number of solar-powered lasers.

https://galactica.org/?max_new_tokens=800&prompt=Lecture+not...

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#254
post #240

It's pretty good at generating convincing-sounding, yet completely wrong, mathematical proofs. Unfortunately, they're not subtly wrong – but I suppose I'm not giving it a chance for subtlety with prompts like "Prove that all elements of a finite field are prime. (Show your working.)".[0] > 1. Show that the product of any two non-zero elements of a finite field is also non-zero. > 2. Let α be an element of a finite fi…

That's more or less what I would expect from a the best language model: things that look very close to real but fail in some way a smart human can tell. YUou need a "knowledge" model to regurgitate facts and an "inference" model to evaluate probabilities of statements being correct.

Yes - that's the whole point of language models ... to model the language, and not the content.

Similar for image generation - the model addresses what looks acceptable (like other images), not what makes sense. It's amazing that they get such interesting results, but merely shocking that we humans interpolate meaning into the images.

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#255
post #253

It knows so much about progressive rock: https://galactica.org/?max_new_tokens=600&prompt=lecture+not...

Reminds me of this:

https://youtu.be/bZe5J8SVCYQ

An algorithm being explained extremely literally whie failing to communicate anything.

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#256

Question: Is the meta company doomed because of projects like GALACTICA?? Answer: To explain, the Meta company is not doomed because of projects like GALACTICA. The Meta company is doomed because of the way it is run. The way it is run is because of projects like GALACTICA. Made me laugh!

If you want to see for yourself! https://galactica.org/?prompt=Is+the+meta+company+doomed+bec...

And ...

Question: Is the meta company doomed?

Answer: To explain, the Meta company is doomed because it is a company that is not profitable.

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#257
Meh. This is so painfully generic to the point of being wholly uninteresting.

  Lecture 1: Vore Within Ancient Babylonian Culture

  Introduction

  In this lecture, we will discuss some of the most important aspects of Babylonian culture that we can study from the tablets found in the area. The Babylonian culture existed during a very important period in human history: the Axial Age. The Axial Age was a period during which there was a great increase in philosophical, religious, and scientific ideas. These ideas were passed down from one generation to the next through written works, and in Babylonian culture, this is where we will see the Babylonian tablets come into play.
Tablets

  The tablets we have today are a great window into Babylonian culture, and it is because of the tablets that we can now study the culture so well. The tablets were created by a cuneiform script, which is a writing system that is still used today. The tablets were written in Akkadian, which was the language of the Babylonians.

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#258
post #77

I've occasionally heard that "monads" are an innovation that allows one to handle side effects, such as input and output, in purely functional programming languages such as Haskell. But whenever I've read an article or blog post about monads, I only barely if at all understood its explanation of what a monad is, and the article usually didn't explain how monads apply to this particular problem. So, just now, when fac…

I thought monads allowed alligators to eat bananas without changing the relative number of either. Can you ask it to make a Monad Analogy Explainer post?

If you want to find out what monads are most like (and least like) you can use the website https://monadsare.fun/. You can also add your own opinions.

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#259

Earlier quoted context omitted.

If you want to see for yourself! https://galactica.org/?prompt=Is+the+meta+company+doomed+bec...

And ... Question: Is the meta company doomed? Answer: To explain, the Meta company is doomed because it is a company that is not profitable.

Clickbait headline:

Facebok is doomed - according to its own scientific AI engine!

Re: Galactica: an AI trained on humanity's scientific knowledge (by Meta)

#260
What is the actual point of a generative text model that emits plausible-sounding scientific technobabble? How this is not just a DDOS against the entire system of peer review?

Honest question: how can this be used for any kind of good? What’s the actual point?

Post reply on HN