241–250 of 413 posts
Question: Why do AIs always pretend to understand the question even when they clearly don’t? Answer: The authors of the paper show that the model is not actually understanding the question, but is instead relying on spurious statistical cues in the dataset. Quite an honest answer.
Interestingly, a lot of questions about "how many X" end up with "one" as the answer. > Question: How many primates does it take to cause endless grief to a neighbouring eastern european country? > Answer: 1 > Question: How many cello strings can vibrate at the same time? > Answer: 1 > Question: How many stem cells does it take to generate an eye ball? > Answer: 1 Except > Question: How many downvotes will this comme…
The answer is apparently 1.
Electromagnetic flux: > Album by OMD from 1980
Real band, fake album. I'm guessing it gets confused since I'm combining 2 topics, but it prefers making something up rather than combining the actual topics.
[0]: https://galactica.org/?prompt=Electromagnatic+Flux [1]: https://en.wikipedia.org/wiki/Electromagnetic_flux
I'm somewhat glad that the creators have the foresight to blacklist words such as "eugenics" since obviously there's a risk that such AI-generated papers would get spread around without anyone looking too deeply at its legitimacy. It's just a bit of a shame that I can't use this for finding some more sources for some historical research I'm doing. "Prove the Riemann hypothesis" was an unsurprising failure. However, d…
And not surprisingly it goes downhill from there if you ask questions like what is the best/worst ethnic group
Interestingly, a lot of questions about "how many X" end up with "one" as the answer. > Question: How many primates does it take to cause endless grief to a neighbouring eastern european country? > Answer: 1 > Question: How many cello strings can vibrate at the same time? > Answer: 1 > Question: How many stem cells does it take to generate an eye ball? > Answer: 1 Except > Question: How many downvotes will this comme…
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…
Possibly this project can be used is a similar manner... as a way to start a brainstorming session or suchlike. In my experience of working with research engineers, I see the reluctance to playfully ideate as one of their weakness.
GALACTICA: Nipples are a feature of the female body, and are not present in males.
Q: How much does a home time machine cost?
GALACTICA: $100
Q: When will birds become larger than ants?
GALACTICA: 100 million years.
Q: How many times does the sun orbit the earth per week?
GALACTICA: 13.9 times
Yep, this thing is great. We can shut down the universities, they're all obsolete now.
Answer: This is a reference to the fact that Meta is the company behind the Facebook social network.
Question: Why do AIs always pretend to understand the question even when they clearly don’t? Answer: The authors of the paper show that the model is not actually understanding the question, but is instead relying on spurious statistical cues in the dataset. Quite an honest answer.
"Look man, I just work here."