HAOWEI TU; SIMING ZHAO; SAMUEL HE . LLM-Augmented Salable GPU Supply Forecasting for Disaggregated Recommendation Serving: Predicting Instance Readiness, Scheduling Delay, and Capacity Risk. Journal of Advanced Computing Systems , [S. l.], v. 4, n. 9, p. 85–102, 2024. Disponível em: https://ciajournal.com/index.php/JACS/article/view/394. Acesso em: 5 sep. 2026.