Ruhr-Uni-Bochum

Yuval Yarom Receives Two Distinguished Paper Awards at CCS 2024

CASA PI Yuval Yarom (Chair for Computer Security) has been awarded two Distinguished Paper Awards at the ACM Conference on Computer and Communications Security (CCS) 2024 in Salt Lake City, USA. The conference took place from 14 – 18 October 2024.

Copyright: RUB, Katja Marquard

CASA PI Yuval Yarom, Professor for Computer Security at the Horst-Görtz-Institute for IT Security, Faculty of Computer Science of the Ruhr-University Bochum, received the first award for the paper Spec-o-Scope: Cache Probing at Cache Speed, co-authored with Gal Horowitz and Eyal Ronen from Tel Aviv University. In this work, the authors investigate a phenomenon called "weird gates", which allows performing arbitrary computation hidden in a computer's microarchitecture. They use the new understanding to develop a new microarchitectural attack called Spec-o-Scope, which is significantly faster than its predecessor and can be used to efficiently recover keys from T-Table AES. They also demonstrate the first attack on unmodified S-Box AES without relying on complicated assumptions.

The second award was given to Yarom for the paper Testing Side-Channel Security of Cryptographic Implementations Against Future Microarchitectures. This work, which also involves CASA PIs Gilles Barthe, Marcel Böhme, and Peter Schwabe (all from the Max Planck Institute for Security and Privacy), as well as Sunjay Cauligi, Daniel Genkin, Marco Guarnieri, David Mateos Romero, and David Wu, explores how future microarchitectural optimizations for speed can compromise the security of existing cryptographic implementations. The team develops a domain-specific language model called LmSpec and a testing framework named LmTest to analyze the effects of 18 proposed optimizations on 25 implementations of cryptographic primitives. The results indicate that all tested implementations exhibit dependencies that could potentially be exploited to decrypt secret keys.

General note: In case of using gender-assigning attributes we include all those who consider themselves in this gender regardless of their own biological sex.