News for hackers. Stuff that matters.
Re:GPT-5.1: A smarter, more conversational ChatGPT by SquareWheel on Thursday November 13, 2025 @11:45AM (#45917082)
Forcing shorter answers will definitely reduce their quality. Every token an LLM generates is like a little bit of extra thinking time. Sometimes it needs to work up to an answer. If you end a response too quickly, such as by demanding one-word answers, it's much more likely to produce hallucinations.
Re:GPT-5.1: A smarter, more conversational ChatGPT by profunctor on Thursday November 13, 2025 @12:31PM (#45917759)
Is this proven?
Re:GPT-5.1: A smarter, more conversational ChatGPT by abustamam on Thursday November 13, 2025 @01:24PM (#45918519)
It's certainly true anecdotally. I've seen it personally plenty of times and I've seen it reported plenty of times.
Re:GPT-5.1: A smarter, more conversational ChatGPT by holbrad on Thursday November 13, 2025 @04:30PM (#45920851)
I know Andrej Karpathy mentions it in his youtube series so there's a good chance of it being true.