
A multi-agent framework that integrates static analysis and dynamic fuzzing to generate secure and functionally correct code by leveraging large language models.


coremsg

securing-llm-generated-code-through-multi-agent-static-analysis-and-dynamic-fuzzing


Securing LLM-Generated Code through Multi-Agent Static Analysis and Dynamic Fuzzing


title_rewrite


Large language models can be enhanced to generate more secure code by automatically synthesizing pairs of vulnerable and fixed code samples, and fine-tuning the models using this data.


enhancing-secure-code-generation-in-large-language-models-via-oracle-guided-synthetic-training-data


Enhancing Secure Code Generation in Large Language Models via Oracle-Guided Synthetic Training Data
