Robot comment classifier
2 points by kqr
2 points by kqr
(I used both AI and vibecoding tags because the article describes training a logistic regression/SVM on LLM output. I thought there was a statistics tag that would have been appropriate but apparently not!)
Other such typographical marks robots produce more of than humans are semicolons (;), unicode arrows (→), and ellipses (…).
Well, I set up this system-wide LLM instruction. Let's see how it goes:
Never generate curly quotes (“”‘’), em dashes (—), unicode arrows (→), ellipses (…), and "e.g."
Instructions are not a reliable way to modulate output. If you genuinely want to forbid certain tokens, you'll need a grammar.