AI RESEARCH
PRBench: A Standardized Probabilistic Robustness Benchmark
arXiv CS.LG
•
ArXi:2511.01724v3 Announce Type: replace-cross Deep learning models are notoriously vulnerable to imperceptible perturbations. Most existing research centers on adversarial robustness (AR), which evaluates models under worst-case scenarios by examining the existence of deterministic adversarial examples (AEs). In contrast, probabilistic robustness (PR) adopts a statistical perspective, measuring the probability that predictions remain correct under stochastic perturbations. While PR is widely regarded as a practical complement to AR, dedicated