While large language models (LLMs) present significant potential for supporting numerous
real-world applications and delivering positive social impacts, they still face significant …

For safety reasons, large language models (LLMs) are trained to refuse harmful user
instructions, such as assisting dangerous activities. We study an open question in this work …

With the continuous development of large language models (LLMs), transformer-based
models have made groundbreaking advances in numerous natural language processing …

Large language models (LLMs) have been rapidly adopted, as showcased by ChatGPT's
overnight popularity, and are integrated in products used by millions of people every day …

Jailbreak attacks in large language models (LLMs) entail inducing the models to generate
content that breaches ethical and legal norm through the use of malicious prompts, posing a …

Metaphor serves as an implicit approach to convey information, while enabling the
generalized comprehension of complex subjects. However, metaphor can potentially be …

Jailbreak prompts pose a significant threat in AI and cybersecurity, as they are crafted to
bypass ethical safeguards in large language models, potentially enabling misuse by …

Black-box jailbreak attacks and defenses, a critical branch of the large language model
(LLM) security, are characterized by their minimal requirement for user expertise and high …