1 paper
Muxi Diao, Rumei Li, Shiyang Liu +4
As large language models (LLMs) continue to advance in capability and influence, ensuring their security and preventing harmful outputs has become crucial. A promising approach to…