Hacker News new | past | comments | ask | show | jobs | submit login

LLMs are known to do bad calculations because all it does is to predict the next token. It’s pretty annoying, and I don’t understand why calculation (eg via Python) is still not a core functionality



Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: