Return to Article Details Energy-Efficient Edge Intelligence through Adaptive Fast–Slow Inference Scheduling in LLM-Driven Systems Download Download PDF