1
SPARK: Security Knowledge Priming and Representation-Guided Knowledge Activation for LLM-based Secure Code Generation
提出SPARK方法，通过安全知识启动和表示引导激活，高效提升大语言模型代码生成的安全性。
arXiv:2606.16244v1 Announce Type: cross Abstract: Large language models routinely generate code with exploitable security flaws. Prior literature attr…