1 articletagged with “based”
Using reinforcement learning to optimize jailbreak strategies against black-box language models.