MARS is a transfer-based black-box attack that uses bi-level optimization on semantic and artifact anchors to escape the linearity trap and improve attack success rates on SSL-SVDD by up to 36%.
Nesterov accelerated gradient and scale invariance for adversarial attacks
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
FogFool creates fog-based adversarial perturbations using Perlin noise optimization to achieve high black-box transferability (83.74% TASR) and robustness to defenses in remote sensing classification.
INTARG generates effective real-time adversarial attacks on time-series regression models by selectively targeting high-confidence high-error steps in a bounded-buffer online setting, increasing prediction error up to 2.42x while attacking under 10% of timesteps.
MCRMO-Attack raises universal targeted attack success rates on unseen images by 23.7% on GPT-4o and 19.9% on Gemini-2.0 over prior universal baselines through stabilized supervision and meta-optimization.
CRHNs integrate convolutional extraction with subspace attractor retrieval trained via Subspace Rotation Algorithm and report order-of-magnitude lower reconstruction error than MHNs and PCNs on STL data under adversarial perturbations.
citing papers explorer
-
Escaping the Linearity Trap: Manifold Detours for Black-Box Adversarial Attacks on Singing Audio Deepfake Detection
MARS is a transfer-based black-box attack that uses bi-level optimization on semantic and artifact anchors to escape the linearity trap and improve attack success rates on SSL-SVDD by up to 36%.
-
Physically-Induced Atmospheric Adversarial Perturbations: Enhancing Transferability and Robustness in Remote Sensing Image Classification
FogFool creates fog-based adversarial perturbations using Perlin noise optimization to achieve high black-box transferability (83.74% TASR) and robustness to defenses in remote sensing classification.
-
INTARG: Informed Real-Time Adversarial Attack Generation for Time-Series Regression
INTARG generates effective real-time adversarial attacks on time-series regression models by selectively targeting high-confidence high-error steps in a bounded-buffer online setting, increasing prediction error up to 2.42x while attacking under 10% of timesteps.
-
Universal Adversarial Attacks against Closed-Source MLLMs via Target-View Routed Meta Optimization
MCRMO-Attack raises universal targeted attack success rates on unseen images by 23.7% on GPT-4o and 19.9% on Gemini-2.0 over prior universal baselines through stabilized supervision and meta-optimization.
-
Robust Auto-associative Memory via Convolutional Restricted Hopfield Networks
CRHNs integrate convolutional extraction with subspace attractor retrieval trained via Subspace Rotation Algorithm and report order-of-magnitude lower reconstruction error than MHNs and PCNs on STL data under adversarial perturbations.