Low-rank additive experts with dual-scale gating and threat-guided diversification improve multi-perturbation adversarial robustness by routing different threat types through distinct model pathways.
In: ICLR (2015)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2representative citing papers
LPID confines perturbations to the face region and optimizes them through a differentiable crop-and-resize model, holding attacker accuracy below 10% on unseen identities where prior unlearnable-example methods collapse to 37–74%.
citing papers explorer
-
RoME: Robust Mixture of Low-Rank Experts against Multiple Adversarial Perturbations
Low-rank additive experts with dual-scale gating and threat-guided diversification improve multi-perturbation adversarial robustness by routing different threat types through distinct model pathways.
-
Unlearnable Faces: Privacy Protection Surviving Extraction Pipeline
LPID confines perturbations to the face region and optimizes them through a differentiable crop-and-resize model, holding attacker accuracy below 10% on unseen identities where prior unlearnable-example methods collapse to 37–74%.