News for hackers. Stuff that matters.
Re:Should We Respect LLMs? A Study on Influence of Prompt Po... by esafak on Tuesday April 22, 2025 @09:12AM (#43761692)
That is not true; first he says it's "tens of millions of dollars well spent," followed by "you never know". I don't think he knows.
Re:Should We Respect LLMs? A Study on Influence of Prompt Po... by lyjackal on Tuesday April 22, 2025 @09:27AM (#43761869)
I’ve wondered whether they use thanks as a signal to a conversation well done, for the purpose of future reinforcement learning
Re:Should We Respect LLMs? A Study on Influence of Prompt Po... by eru on Tuesday April 22, 2025 @09:33AM (#43761960)
I'd speculate that they use slightly more complicated sentiment analysis. This has been a thing since long before LLMs.
Re:Should We Respect LLMs? A Study on Influence of Prompt Po... by esafak on Tuesday April 22, 2025 @09:32AM (#43761935)
I don't know if they do or if it is efficient but it is possible.