{"data":{"id":"3f06df13-6c62-4e66-b320-37c672f7cc3e","title":"Assessing and Improving DNN Robustness Against Adversarial Examples From the Perspective of Fully Connected Layers","summary":"Researchers propose a redundant fully connected layer that can be plugged into existing image classification model backbones to improve adversarial robustness. The layer is trained with a loss function that uses cosine similarity to maximize the difference and diversity among multiple fully connected parts. Tested against 12 representative white-box and black-box attacks on two popular datasets, the approach reportedly gives significant robustness with negligible extra training overhead and hardly degrades clean-sample accuracy.","solution":"The proposed mitigation is the redundant fully connected layer, combined with existing model backbones in a pluggable manner, trained with the cosine-similarity-based loss function described in the source.","labels":["security","research"],"sourceUrl":"http://ieeexplore.ieee.org/document/11456181","publishedAt":"2026-03-25T13:17:12.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":["model_evasion"],"issueType":"research","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":[],"affectedVendorsRaw":[],"classifierModel":"claude-haiku-5-5","classifierPromptVersion":"v4","summaryPromptVersion":"v2","headline":null,"headlinePromptVersion":null,"cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"epssCheckedAt":null,"kevDateAdded":null,"advisoryAliases":null,"affectedPackagesSource":null,"affectedPackagesCheckedAt":null,"patchAvailable":null,"disclosureDate":"2026-03-25T13:17:12.000Z","capecIds":null,"crossRefCount":0,"attackSophistication":"advanced","impactType":["integrity"],"aiComponentTargeted":"model","llmSpecific":false,"classifierConfidence":0.95,"researchCategory":"peer_reviewed","atlasIds":null}}