Makes me wonder if "I don't know" could be added to LLM: whenever an activation ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		yukIttEft on Sept 13, 2024 \| parent \| context \| favorite \| on: Notes on OpenAI's new o1 chain-of-thought models Makes me wonder if "I don't know" could be added to LLM: whenever an activation has no clear winner value (layman here), couldn't this indicate low response quality?

Regic on Sept 14, 2024 [–]

This exists and does work to some degree, e.g. Detecting hallucinations in large language models using semantic entropy https://www.nature.com/articles/s41586-024-07421-0

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact