Follow Slashdot stories on Twitter

 



Forgot your password?
typodupeerror

Comment Re:LLMs predict (Score 1) 238

what kind of behavior would demonstrate that LLMs did have understanding?

An LLM would need to act like an understander -- the essence of the Turing Test. Exactly what that means is a complex question. And it's a necessary but not sufficient condition. But we can easily provide counterexamples where the LLM is clearly not an understander. Like this from the paper:

When prompted with the CoT prefix, the modern LLM Gemini responded: âoeThe United States was established in 1776. 1776 is divisible by 4, but itâ(TM)s not a century year, so itâ(TM)s a leap year. Therefore, the day the US was established was in a normal year.â This response exemplifies a concerning pattern: the model correctly recites the leap year rule and articulates intermediate reasoning steps, yet produces a logically inconsistent conclusion (i.e., asserting 1776 is both a leap year and a normal year).

Comment Re:Company selling (Score 1) 168

I've been asked to create reports that add pounds + gallons, and it's almost impossible to get them to understand why that's nonsense.

Pshaw, that's super easy! 3 pounds plus 6 gallons equals 9.

Perhaps I'm being too harsh. For example, if their boss is an MBA who gives out raises on the basis of how many pounds+gallons they produce or sell, they would be quite rational to request a report that shows how many pounds+gallons they have produced or sold.

Comment Re:Do religion next! (Score 1) 111

how is that not also fraud?

I've been wondering what the authorities would do if I started selling updated accommodations for the afterlife. Want an extra garage or bath? Something closer to the golden throne, or further from all that off-key singing? Or most popular of all, something farther from those people.

Pay now and get it later, of course.

Slashdot Top Deals

He: Let's end it all, bequeathin' our brains to science. She: What?!? Science got enough trouble with their OWN brains. -- Walt Kelly

Working...