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

Perplexity is a term of art in LLM training, yes.

A naive way of scoring how AI laden text is would be to run n-1 layers of a model and compare the text to the probability space of tokens from the model.

It works somewhat to detect obvious text but is not strong enough a method by itself.



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

Search: