“Reducing Latency and Enhancing Accuracy in LLM Inference through Firmware-Level Optimization”, ijvsli, vol. 5, no. 02, pp. 26–36, Jul. 2025, doi: 10.55640/ijvsli-05-02-02.