Abstract: Large language models (LLMs) demonstrate promising capabilities for automated security vulnerability detection, yet current evaluation methodologies lack statistical validation to ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results