This research introduces an innovative approach to wireless communication that prioritizes meaning over raw data transmission, significantly improving efficiency for bandwidth-intensive applications like AR/VR.

• Integrates Large Language Models with knowledge graphs to create an end-to-end semantic communication system
• Reduces transmission overhead by sending only the most meaningful information
• Demonstrates superior performance compared to traditional bit-transmission approaches
• Provides a scalable solution for emerging high-bandwidth applications

For the engineering community, this research offers a promising pathway beyond Shannon theory limitations, potentially transforming how we design next-generation wireless networks to handle exponentially growing data demands.

LLM-Enabled Data Transmission in End-to-End Semantic Communication