Paper analyzing the architecture and performance of RAG systems accepted to ISCA 2025

The paper “RAGO: A Systematic Framework for Design and Optimization of Retrieval-Augmented Generation Serving” by Wenqi Jiang (Google), Suvinay Subramanian (Google), Cat Graves (Google), Gustavo Alonso (ETH Zurich), Amir Yazdanbakhsh (Google DeepMind), and Vidushi Dadu (Google) has been accepted for publication at the external page International Symposium on Computer Architecture (ISCA) that will take place in June, 2025 in Tokyo, Japan.