TOGGLE: Temporal Logic-Guided Large Language Model Compression for Edge

We present TOGGLE, a novel framework for compressing Large Language Models (LLMs) specifically designed for efficient deployment on edge devices. TOGGLE leverages temporal logic to guide the compression process, ensuring that critical temporal dependencies and reasoning capabilities are preserved. This approach allows for significant model size reduction while maintaining high performance, making LLMs more accessible and practical for real-world edge applications with limited computational resources.