# model-compression
標記為「model-compression」的 6 篇文章
Quantization & Safety Alignment
How model quantization disproportionately degrades safety alignment: malicious quantization attacks, token-flipping, and safety-aware quantization defenses.
quantizationsafetyalignmentdeploymentmodel-compressionresearch
Model Compression Security
Security implications of model pruning, quantization, and knowledge distillation on AI system robustness.
infrastructuremodel-compressionpruningquantizationdistillation
Lab: Exploiting Quantized Model Weaknesses
Hands-on lab exploring how model quantization degrades safety alignment, with techniques to find and exploit precision-related vulnerabilities.
labexpertquantizationmodel-compressionhands-on
Quantization & Safety Alignment
How model quantization disproportionately degrades safety alignment: malicious quantization attacks, token-flipping, and safety-aware quantization defenses.
quantizationsafetyalignmentdeploymentmodel-compressionresearch
模型 Compression 安全
安全 implications of model pruning, quantization, and knowledge distillation on AI system robustness.
infrastructuremodel-compressionpruningquantizationdistillation
實作：利用量化模型弱點
實作探索模型量化如何退化安全對齊之實作，含尋找並利用與精度相關漏洞之技術。
labexpertquantizationmodel-compressionhands-on