Paper says target agents criticize attacks better than they propose them
Separating criticism and proposal reached 63.2% attack success on DTAP-BENCH across four domains.
Stories per day · 30 days
1 stories, newest first · one card per story
1 stories, newest first · one card per story
Separating criticism and proposal reached 63.2% attack success on DTAP-BENCH across four domains.
Today 1.2K readers · 1.8K views · 1.4K reads · all time 160K views