LLMs Cannot Reliably Identify and Reason about Security Vulnerabilities (Yet?): A Comprehensive Evaluation, Framework, and BenchmarksS. Ullah, M. Han, S. Pujar, H. Pearce, A. Coskun, G. StringhiniJan 1, 2024DOI URL