Large language models can acquire undesirable capabilities during pretraining that complicate model deployment. Machine unlearning offers one approach to this challenge by attempting to remove these ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results