Side-Channel Papers Still Benchmark Against Proxies Set by Early Attack Work, Including at Top-Tier Venues

arXiv 2609.03893·low signalXBlueskyLinkedInCopy link

Microarchitectural side-channel research grew fast on de facto benchmarks inherited from early attack papers, which relied on indirect proxies such as covert-channel bandwidth or key recovery against naive AES and RSA implementations, and later work kept comparing against raw numbers lifted from prior papers. Because these attacks are acutely sensitive to experimental conditions, where small target-system changes shift outcomes substantially, the authors argue the practice undermines reproducibility and makes comparisons doubtful even where peer review should have caught it. The paper takes on proper benchmarking for this class of attack directly rather than proposing another primitive.