Psychology
fromLesswrong
5 days agoA Mirror Test For LLMs - LessWrong
A new measure of LLM self-awareness is proposed, but current models ultimately fall short in demonstrating true self-awareness.
"It's definitely a conversation starter," said Snyder, a three-time world Sudoku champion and an author and editor of more than dozens of ebooks such as 'The Art of Sudoku' and 'The Art of Puzzles.'"
Computational linguistics is a two-way street: You're either using a computer to do things with human language or communicate or translate or teach a foreign language, or you're using computational techniques to learn something about human languages. Her work documenting and preserving endangered languages uses a little bit of both.
With its Alpha series of game-playing AIs, Google's DeepMind group seemed to have found a way for its AIs to tackle any game, mastering games like chess and by repeatedly playing itself during training. But then some odd things happened as people started identifying Go positions that would lose against relative newcomers to the game but easily defeat a similar Go-playing AI.
Liam Delap has an interesting party trick that you wouldn't really expect a footballer to pull out of the bag. The Chelsea forward has gone on video a few times showing off his bizarre mathematical ability to quickly calculate the cube roots of large numbers.
In Texas Hold'em poker, players wager on the best five-card hand they can make among the two cards in their hand and the communal ones on the table. Hands are ranked based on their probability of occurring. A full house, for example, with three cards of the same value (fives or kings, for instance) and two cards of another, is less likely than a flush with any five cards of the same suit. A full house therefore beats a flush.
Which Algorithm Is This? If you step back, this maps almost perfectly to the Top K Frequent Elements problem.We usually solve it for integers in a list. Here, the "elements" are audience profiles age and body-type combinations. First, define what an audience profile looks like: case class Profile(age: Int, height: Int, weight: Int) What we want is a function like this:
In the weeks leading up to September 1891, mathematician Georg Cantor prepared an ambush. For years he had sparred - philosophically, mathematically and emotionally - with his formidable rival Leopold Kronecker, one of Germany's most influential mathematicians. Kronecker thought that mathematics should deal only with whole numbers and proofs built from them and therefore rejected Cantor's study of infinity. "God made the integers," Kronecker once said. "All else is the work of man."
Walking through a field one day, a 17-year-old schoolteacher named George Boole had a vision. His head was full of abstract mathematics - ideas about how to use algebra to solve complex calculus problems. Suddenly, he was struck with a flash of insight: that thought itself might be expressed in algebraic form. Boole was born on November 2, 1815, at four o'clock in the afternoon, in Lincoln, England.
You know that sinking feeling when you realize you've been using a phrase that makes you sound less intelligent than you actually are? I had one of those moments a few years back during a pitch meeting for my startup. I was presenting to potential investors, and I kept saying "I think" before every point I made. "I think our user acquisition strategy will work."
This week's puzzle was constructed by Rebecca Goldstein and Kelsey Dixon, and edited by Hoang-Kim Vu. Rebecca is a crossword constructor from the Bay Area, and Kelsey is a crossword constructor from Chicago. They both lived in Atlanta in the '90s, which is why Kelsey has been trying to start a rumor that Rebecca was her childhood babysitter. They hope you don't take the puzzle too seriously!
It may seem like they've been around forever, but the crossword as we know it is barely a century old. They started in the New York World in 1913, where it was originally called a "word-cross." Going on to obsess writers like T.S. Eliot and Vladimir Nabokov, who reportedly wrote the first Russian-language puzzle as a teenager, the crossword settled into a kind of urbane normalcy over the course of the 20th century, a feature of newspapers and cheap jumbo packs.
On October 1, 2022, something strange happened in the Philippines: 433 people won the jackpot in the local lottery. For this particular lotto, six numbers ranging in value from 1 to 55 were randomly selected, and the 433 winners all matched. Even more bizarre, when arranged in ascending order, the winning numbers were: 9, 18, 27, 36, 45 and 54. In other words, the winning numbers were multiples of 9 (9 1, 9 2, 9 3, etcetera).
I'm going to give you some clues. The answer to each one rhymes with the last word in the clue. Ex. The sky's hue --> Blue 1. Toy that flies to great height 2. Pistol, for one 3. Funeral fire 4. Things you count when you have trouble getting to sleep 5. Friars event with a celebrity host 6. Brand of pen that you can click 7. Place to acquire knowledge 8. Have uncertainty about 9. Not go away
A drawn circle is at least something physical. You can see it, touch it, erase it. The skeptic can still say, "Circles are grounded in physical reality. Justice is different; it's just an idea in your head." So let's talk about the number two. Point to it. Not two apples, not two fingers, not a numeral on a page-that's just a symbol.
In January 1986, NASA engineers knew the Space Shuttle Challenger's O-rings had never been tested in freezing temperatures. They recommended delaying the launch. Managers asked: Could the engineers prove it was unsafe? They couldn't-they could only say the system hadn't been designed for these conditions. Under pressure, the engineers withdrew their recommendation. The next morning, Challenger broke apart 73 seconds after launch, killing all seven astronauts.