Sensitivity as a Double-Edged Sword: A Trade-off Between Discriminability and Adversarial Robustness

• URL: http://arxiv.org/abs/2606.01746v1
• Date: Mon, 01 Jun 2026 06:09:19 GMT
• Title: Sensitivity as a Double-Edged Sword: A Trade-off Between Discriminability and Adversarial Robustness
• Abstract summary: Modern neural networks are highly susceptible to adversarial perturbations.<n>In this work, we identify that part of this vulnerability stems from the sensitivity of the widely used fully connected (FC) classifiers to such perturbations.<n>We propose a new, rigorous evaluation protocol, the Mixed Surrogate Attack (MSA), which uses multiple surrogates along with powerful AutoAttack to ensure a fair and robust assessment.
• Score: 4.214948855787466
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modern neural networks are highly susceptible to adversarial perturbations. In this work, we identify that part of this vulnerability stems from the sensitivity of the widely used fully connected (FC) classifiers to such perturbations. In contrast, simple $\ell_2$ distance-based classifiers exhibit significantly greater robustness. We provide thorough theoretical and empirical analysis showing that while FC classifiers' high sensitivity makes them discriminative, it also makes them vulnerable. Conversely, $\ell_2$-classifiers' insensitivity grants robustness but limits performance. Motivated by this trade-off, we propose a novel $\ell_2$-reclassifier based on a Hybrid Prototype Mixing (HPM) framework. This method retains the discriminative power of FC classifiers while leveraging the robustness of $\ell_2$ distance. It yields $\ell_2$-distance-based predictions by fusing two prototype types: (1) stable, dataset-level prototypes updated via EMA, and (2) dynamic, batch-level prototypes generated from the FC classifier's predictions using a Straight-Through Estimator (STE). However, this dynamic, STE-based architecture introduces significant challenges for evaluation, such as gradient obfuscation and forward discontinuity. To address this, we propose a new, rigorous evaluation protocol, the Mixed Surrogate Attack (MSA), which uses multiple surrogates along with powerful AutoAttack to ensure a fair and robust assessment. Extensive experiments demonstrate that our lightweight, plug-and-play module, with minimal fine-tuning, effectively enhances the adversarial robustness of various existing SOTA adversarially trained models.

Related papers

• ARMOR++: Agentic Orchestration of a Multi-Domain Primitive Set for Transferable Attacks on Deepfake Detectors [28.44035744358622]
  This paper introduces ARMOR++, a robust multi-agent framework designed for high-transferability deepfake evasion.<n>By integrating five complementary primitives, ARMOR++ effectively targets heterogeneous inductive biases.<n> Rigorous evaluation on the AADD-2025 benchmark demonstrates that ARMOR++ significantly outperforms existing agentic and non-agentic baselines.
  arXiv  Detail & Related papers  (2026-07-16T17:45:13Z)
• TaFD: Threat-Aware Frequency Decoupling for Adversarial Robustness against Heterogeneous Attacks [6.525195905471157]
  Multi-threat robustness remains a fundamental challenge in deep learning.<n>Joint adversarial training (JAT) is widely adopted, but suffers from negative transfer under heterogeneous threats.<n>We propose Threat-aware Frequency Decoupling (TaFD), a two-stage defense framework that reformulates JAT as a frequency-domain divide-and-conquer paradigm.
  arXiv  Detail & Related papers  (2026-06-16T05:38:26Z)
• NetDiffuser: Deceiving DNN-Based Network Attack Detection Systems with Diffusion-Generated Adversarial Traffic [1.8225278540778656]
  This paper proposes NetDiffuser, a framework for generating natural adversarial examples (NAEs) capable of deceiving Network Intrusion Detection System (NIDS)<n>NetDiffuser achieves up to a 29.93% higher attack success rate and reduces AE detection performance by at least 0.267 (in some cases up to 0.534) in the Area under the Receiver Operating Characteristic Curve (AUC-ROC) score compared to the baseline attacks.
  arXiv  Detail & Related papers  (2026-03-09T20:13:51Z)
• GuardFed: A Trustworthy Federated Learning Framework Against Dual-Facet Attacks [56.983319121358555]
  Federated learning (FL) enables privacy-preserving collaborative model training but remains vulnerable to adversarial behaviors.<n>We introduce the Dual-Facet Attack (DFA), a novel threat model that concurrently undermines predictive accuracy and group fairness.<n>We propose GuardFed, a self-adaptive defense framework that maintains a fairness-aware reference model using a small amount of clean server data.
  arXiv  Detail & Related papers  (2025-11-12T13:02:45Z)
• Deep Positive-Negative Prototypes for Adversarially Robust Discriminative Prototypical Learning [0.24999074238880484]
  We propose a novel framework named Adversarially trained Deep Positive-Negative Prototypes (Adv-DPNP)<n>Adv-DPNP integrates discriminative prototype-based learning with adversarial training.<n>We show that Adv-DPNP achieves the highest average accuracy across severities and corruption types.
  arXiv  Detail & Related papers  (2025-04-03T15:42:58Z)
• Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria [61.048842737581865]
  We show that Adversarial Training (AT) omits to learning robust features, resulting in poor performance of adversarial robustness. We propose a generic framework of AT to gain robust representation, by the asymmetric negative contrast and reverse attention. Empirical evaluations on three benchmark datasets show our methods greatly advance the robustness of AT and achieve state-of-the-art performance.
  arXiv  Detail & Related papers  (2023-10-05T07:29:29Z)
• Improving Adversarial Robustness to Sensitivity and Invariance Attacks with Deep Metric Learning [80.21709045433096]
  A standard method in adversarial robustness assumes a framework to defend against samples crafted by minimally perturbing a sample. We use metric learning to frame adversarial regularization as an optimal transport problem. Our preliminary results indicate that regularizing over invariant perturbations in our framework improves both invariant and sensitivity defense.
  arXiv  Detail & Related papers  (2022-11-04T13:54:02Z)
• Generative Adversarial Network-Driven Detection of Adversarial Tasks in Mobile Crowdsensing [5.675436513661266]
  Crowdsensing systems are vulnerable to various attacks as they build on non-dedicated and ubiquitous properties. Previous works suggest that GAN-based attacks exhibit more crucial devastation than empirically designed attack samples. This paper aims to detect intelligently designed illegitimate sensing service requests by integrating a GAN-based model.
  arXiv  Detail & Related papers  (2022-02-16T00:23:25Z)
• SmoothMix: Training Confidence-calibrated Smoothed Classifiers for Certified Robustness [61.212486108346695]
  We propose a training scheme, coined SmoothMix, to control the robustness of smoothed classifiers via self-mixup. The proposed procedure effectively identifies over-confident, near off-class samples as a cause of limited robustness. Our experimental results demonstrate that the proposed method can significantly improve the certified $ell$-robustness of smoothed classifiers.
  arXiv  Detail & Related papers  (2021-11-17T18:20:59Z)
• Adaptive Feature Alignment for Adversarial Training [56.17654691470554]
  CNNs are typically vulnerable to adversarial attacks, which pose a threat to security-sensitive applications. We propose the adaptive feature alignment (AFA) to generate features of arbitrary attacking strengths. Our method is trained to automatically align features of arbitrary attacking strength.
  arXiv  Detail & Related papers  (2021-05-31T17:01:05Z)
• Robust Classification Under $\ell_0$ Attack for the Gaussian Mixture Model [39.414875342234204]
  We develop a novel classification algorithm called FilTrun that has two main modules: filtration and Truncation. We discuss several examples that illustrate interesting behaviors such as a phase transition for adversary's budget determining whether the effect of adversarial perturbation can be fully neutralized.
  arXiv  Detail & Related papers  (2021-04-05T23:31:25Z)
• Consistency Regularization for Certified Robustness of Smoothed Classifiers [89.72878906950208]
  A recent technique of randomized smoothing has shown that the worst-case $ell$-robustness can be transformed into the average-case robustness. We found that the trade-off between accuracy and certified robustness of smoothed classifiers can be greatly controlled by simply regularizing the prediction consistency over noise.
  arXiv  Detail & Related papers  (2020-06-07T06:57:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.