Assessing Safety Risks and Quantization-aware Safety Patching for Quantized Large Language Models | Read Paper on Bytez