AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

CERTIFIED VS. EMPIRICAL ADVERSARIAL ROBUSTNESS VIA HYBRID CONVOLUTIONS WITH ATTENTION STOCHASTICITY

Joy Dhar, Song Xia, Manish Kumar Pandey, Maryam Haghighat, Azadeh Alavi, Ferdous Sohel, Wenyu Zhang, Nayyar Zaidi

ICLR 2026 · Adversarial robustness and security › Certified and empirical robustness · approach: Robustness-accuracy trade-off analysis

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase smoothing noise sigma from 0.25 to 0.50 in HyCAS (ablation)HyCAS · CIFAR-10, radius 2.0certified accuracy (%): 8.52 -> 12.5Improved
increase smoothing noise sigma from 0.25 to 0.50 in HyCAS (ablation)HyCAS · ImageNet, radius 2.0certified accuracy (%): 5.42 -> 24.8Improved
increase smoothing noise sigma from 0.25 to 0.50 in HyCAS (ablation)HyCAS · CIFAR-10, radius 0.75certified accuracy (%): 44.3 -> 44.3No clear effect
use HyCAS instead of ARS (hybrid Lipschitz convolutions plus attention noise)randomized smoothing certification · CIFAR-10, radius 0.75, sigma 0.25certified accuracy (%): 39.1 -> 44.3Improved
use HyCAS instead of CTRWempirical l-inf defense · NIH-CXR, APGD-20 eps 8/255robust accuracy (%): 85.1 -> 88.6Improved
use HyCAS instead of CTRWempirical l-inf defense · HAM10000, APGD-20 eps 8/255robust accuracy (%): 64.7 -> 67.8Improved
use HyCAS instead of CTRWempirical l-inf defense · EyePACS, APGD-20 eps 8/255robust accuracy (%): 70.1 -> 72.6Improved
increase APGD attack steps from 10 to 100HyCAS vs prior defenses · CIFAR-100, APGD stepsrobust accuracy: HyCAS stays 7-12% above closest competitor at 100 stepsMixed
increase certified radius (via larger smoothing noise)HyCAS · CIFAR-10 and ImageNetempirical robustnessHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy