One Leak Away: How Pretrained Model Exposure Amplifies Jailbreak Risks in Finetuned LLMs

Finetuning pretrained large language models LLMs has become the standard paradigm for developing downstream applications. However, its security implications remain unclear, particularly regarding whether finetuned LLMs inherit jailbreak vulnerabilities from their pretrained sources. We investigat...

Unlearning Isn'T Deletion: Investigating Reversibility of Machine Unlearning in LLMs

Unlearning in large language models LLMs is intended to remove the influence of specific data, yet current evaluations rely heavily on token-level metrics such as accuracy and perplexity. We show that these metrics can be misleading: models often appear to forget, but their original behavior can ...