Revealing a Hidden Class of Task-in-Prompt Adversarial Attacks on LLMs (aclanthology.org) 1 points by aghuang 7d ago ↗ HN
1 comment
[ 0.26 ms ] story [ 7.4 ms ] thread