This research introduces an innovative LLM-empowered framework that addresses class imbalance challenges in detecting online drug trafficking.

• Leverages graph prompt learning to identify drug trafficking activities on major online platforms
• Overcomes the practical challenge of highly imbalanced data with limited labeled samples
• Demonstrates superior performance compared to traditional detection methods
• Provides a scalable solution for law enforcement to monitor illegal drug sales

Security Impact: This approach offers security professionals and law enforcement agencies a practical tool to identify and combat illegal drug trafficking operations that increasingly use mainstream online platforms as distribution channels.

LLM-Empowered Class Imbalanced Graph Prompt Learning for Online Drug Trafficking Detection