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

LLMs are very good at logical reasoning in bounded systems. They lack the wisdom to deal with unbounded systems efficiently, because they don't have a good sense of what they don't know or good priors on the distribution of the unexpected. I expect this will be very difficult to RL in.
 help



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

Search: