[1]
Lee Ji-su, “GPU Memory Usage Prediction for Generative AI Serving Pipelines with Queue, Latency, and Utilization Signals”, JACS, vol. 6, no. 7, pp. 1–15, Jul. 2026, Accessed: Sep. 06, 2026. [Online]. Available: https://ciajournal.com/index.php/JACS/article/view/422