CIFAR-100

CIFAR-100 is a dataset that contains 100 classes of images, each with 600 images, used to evaluate the performance of machine learning models, particularly in the context of image classification tasks.

Papers using CIFAR-100 (16)

ADS-C: Antidistillation Sampling for Classification2026Toward a Generalized Defense Across Sparse, Continuous, and Structured Parameter Attacks2026Reinforcement Learning Disrupts Gradient-Based Adversarial Optimization2026Adversarial Sparse Teacher: Defense Against Distillation-based Model Stealing Attacks Using Adversarial Examples2024 · 1 citesPubdef: Defending Against Transfer Attacks From Public Models2023DCInject: Persistent Backdoor Attacks via Frequency Manipulation in Personal Federated Learning2026SoK: The Last Line of Defense: On Backdoor Defense Evaluation2025Enhancing All-to-X Backdoor Attacks with Optimized Target Class Mapping2025Variational Autoencoder-Based Black-Box Adversarial Attack on Collaborative DNN Inference2025DASH: A Meta-Attack Framework for Synthesizing Effective and Stealthy Adversarial Examples2025HASSLE: A Self-Supervised Learning Enhanced Hijacking Attack on Vertical Federated Learning2025Smoothed Inference For Adversarially-trained Models2019 · 1 citesEfficient Passive Membership Inference Attack In Federated Learning2021 · 15 citesMembership Inference Attack Using Self Influence Functions2022 · 3 citesAdvancing Adversarial Robustness Through Adversarial Logit Update2023Towards Adversarially Robust Continual Learning2023 · 8 cites