Execution-Based Evaluation of Natural Language to Bash and PowerShell for Incident Remediation

Ngoc Phuoc An Vo, Brent Paulovicks, Vadim Sheinin

Given recent advancements of Large Language Models (LLMs), code generation tasks attract immense attention for wide application in different domains. In an effort to evaluate and s…