News for hackers. Stuff that matters.
Re:Detecting when LLMs are uncertain by astrange on Friday October 25, 2024 @09:30PM (#41951762)
That's just autocorrect. (Or generative AI.)
Re:Detecting when LLMs are uncertain by TeMPOraL on Saturday October 26, 2024 @03:52AM (#41953280)
Or error correction. Or statistical analysis.

"Right" and "wrong" aren't binary states. In many cases, if the data is at least in small part correct, that small part can be used to improve correctness in an automated way.

Re:Detecting when LLMs are uncertain by adrian_b on Saturday October 26, 2024 @04:52AM (#41953531)
Except that autocorrect is frequently wrong, so that many authors of hilariously wrong messages have to apologize that the messages must have been messed by autocorrect (which may be true or not).

When autocorrect is wrong, it usually is because it chooses words believed to be used more frequently in that context, so especially the authors of scientific or technical texts are affected by the wrong guesses of autocorrect, because they use less common words.