News for hackers. Stuff that matters.
Re:Anthropic Claude 3.5 can create icalendar files, so I did... by wanderingmind on Sunday August 25, 2024 @07:53AM (#41346634)
Kind of a noob question, is it possible to design a GAN type network with LLM, where one (or many) LLMs generate outputs, while a few other LLMs validate or discriminate them and thus improving generator LLMs accuracy.
Re:Anthropic Claude 3.5 can create icalendar files, so I did... by oneshtein on Sunday August 25, 2024 @12:36PM (#41348661)
Yes, you can use AI to spot errors in AI output. Done before with good results, but it requires to run 2 different, but equally good, AI models in parallel, which is way more expensive than 1 model.
Re:Anthropic Claude 3.5 can create icalendar files, so I did... by ada1981 on Sunday August 25, 2024 @12:42PM (#41348726)
"which is way more expensive than 1 model."

In this case "way more" means exactly 2x the cost.