Hacker News new | ask | show | jobs
by fiso64 1121 days ago
Humans make foolish, nonsensical mistakes too. Does that mean that they're not intelligent and not sensitive to the meaning of the code?
1 comments
YeGoblynQueenne 1121 days ago
We already know that humans are sensitive to meaning. That's not a hypothesis that needs testing.
link
TeMPOraL 1121 days ago
This experiment only proves LLMs are sensitive to meaning too. It was also already known.
link
YeGoblynQueenne 1121 days ago
All I can see with any clarity is that LLMs are sensitive to the user's prompt.
link