Our work received HPCA 2026 Distinguished Artifact Award

We are excited to share that our work, “Athena: Synergizing Data Prefetching and Off-Chip Prediction via Online Reinforcement Learning,” has been recognized with the Distinguished Artifact Award at the 32nd International Symposium on High-Performance Computer Architecture (HPCA) 2026 in Sydney, Australia.


Huge thanks to all the co-authors: Caroline Hengartner, Konstantinos Kanellopoulos, Rakesh Kumar, Mohammad Sadrosadati, and Onur Mutlu, and especially to my co-primary author, Rahul Bera. This work would not have been possible without his relentless effort, technical depth, and countless discussions that shaped Athena into what it is today.

Through our artifact, we open-sourced a comprehensive collection of prefetchers, off-chip predictors, and coordination policies evaluated in the paper. We hope it will serve as a foundation for future research on learning-driven microarchitectural coordination and reproducible systems research.


Paper: https://ieeexplore.ieee.org/abstract/document/11408449

Artifact: https://github.com/CMU-SAFARI/Athena