Be Kind, Rewrite: Benign Projections Via Rewriting Defend against LLM Data Poisoning Attacks

Large language models LLMs are highly susceptible to backdoor attacks BAs, wherein training samples are poisoned using trigger-based harmful content. Furthermore, existing defenses have proven ineffective when extensively tested across BA patterns. To better combat BAs, we explore the use of LLM...

Behind the Mask: Benchmarking Camouflaged Jailbreaks in Large Language Models

Large Language Models LLMs are increasingly vulnerable to a sophisticated form of adversarial prompting known as camouflaged jailbreaking. This method embeds malicious intent within seemingly benign language to evade existing safety mechanisms. Unlike overt attacks, these subtle prompts exploit...

Exploring the Secondary Risks of Large Language Models

Ensuring the safety and alignment of Large Language Models is a significant challenge with their growing integration into critical applications and societal functions. While prior research has primarily focused on jailbreak attacks, less attention has been given to non-adversarial failures that...