MOMAT: Mixture of Multiple Atlases for Low-Power Jailbreak Defense of Quantized LLMs
Quantized large language models are increasingly deployed on edge devices for their low latency and energy efficiency. However, model quantization weakens alignment safeguards, leaving qLLMs (quantized large language models) highly vulnerable to jailbreak attacks. To address this challenge, we present MOMAT (Mixture of...
Bo-Yang Li, Bingyu Shen, Wei-Hao Hong et al.
· 0 citations