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

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

14 December 2025

Yixin Tan

Zhe Yu

Jun Sakuma

ArXiv (abs)PDF HTML Github (1498★)

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

0 / 0 papers shown

Title
No papers found