CoDet-M4 is a novel framework that accurately identifies machine-generated code across multiple programming languages, LLM generators, and coding domains.

• Addresses critical gaps in code detection by supporting diverse programming languages beyond just Python
• Demonstrates robustness against multiple LLM code generators including ChatGPT, Claude, and others
• Maintains high detection accuracy across various coding domains from algorithms to web development
• Provides a security-focused solution that helps maintain programming integrity, educational standards, and prevents potential misuse

This research is vital for security professionals who need reliable tools to distinguish between human and AI-written code, helping organizations maintain accountability and protect against unauthorized or unattributed code contributions.

CoDet-M4: Detecting Machine-Generated Code in Multi-Lingual, Multi-Generator and Multi-Domain Settings