Image de couverture de livre

Layer-Aware Analysis of Catastrophic Overfitting: Revealing the Pseudo-Robust Shortcut Dependency

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Lin, Runqi, Yu, Chaojian, Han, Bo, Su, Hang, Liu, Tongliang
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

On the Over-Memorization During Natural, Robust and Catastrophic Overfitting
par: Lin, Runqi, et autres
Publié: (2023)

Eliminating Catastrophic Overfitting Via Abnormal Adversarial Examples Regularization
par: Lin, Runqi, et autres
Publié: (2024)

Understanding Robust Overfitting from the Feature Generalization Perspective
par: Yu, Chaojian, et autres
Publié: (2023)

Instance-dependent Early Stopping
par: Yuan, Suqin, et autres
Publié: (2025)

Time-Efficient Evaluation and Enhancement of Adversarial Robustness in Deep Neural Networks
par: Lin, Runqi
Publié: (2025)

Catastrophic Overfitting: A Potential Blessing in Disguise
par: Zhao, Mengnan, et autres
Publié: (2024)

Residual Stream Analysis of Overfitting And Structural Disruptions
par: Liu, Quan, et autres
Publié: (2026)

Unveiling the Backdoor Mechanism Hidden Behind Catastrophic Overfitting in Fast Adversarial Training
par: Zhao, Mengnan, et autres
Publié: (2026)

Benign, Tempered, or Catastrophic: A Taxonomy of Overfitting
par: Mallinar, Neil, et autres
Publié: (2022)

Understanding and Enhancing the Transferability of Jailbreaking Attacks
par: Lin, Runqi, et autres
Publié: (2025)

Risk Phase Transitions in Spiked Regression: Alignment Driven Benign and Catastrophic Overfitting
par: Li, Jiping, et autres
Publié: (2025)

FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction
par: Lin, Runqi, et autres
Publié: (2025)

Theoretical Analysis of Robust Overfitting for Wide DNNs: An NTK Approach
par: Fu, Shaopeng, et autres
Publié: (2023)

Robust Training of Federated Models with Extremely Label Deficiency
par: Zhang, Yonggang, et autres
Publié: (2024)

ShortcutProbe: Probing Prediction Shortcuts for Learning Robust Models
par: Zheng, Guangtao, et autres
Publié: (2025)

Stealthy and Adjustable Text-Guided Backdoor Attacks on Multimodal Pretrained Models
par: Zhang, Yiyang, et autres
Publié: (2026)

MOKD: Cross-domain Finetuning for Few-shot Classification via Maximizing Optimized Kernel Dependence
par: Tian, Hongduan, et autres
Publié: (2024)

Overfitting in Adaptive Robust Optimization
par: Zhu, Karl, et autres
Publié: (2025)

Rethinking Benign Overfitting in Two-Layer Neural Networks
par: Xu, Ruichen, et autres
Publié: (2025)

Enhancing Sample Selection Against Label Noise by Cutting Mislabeled Easy Examples
par: Yuan, Suqin, et autres
Publié: (2025)

Preventing Catastrophic Overfitting in Fast Adversarial Training: A Bi-level Optimization Perspective
par: Wang, Zhaoxin, et autres
Publié: (2024)

Generative Model Inversion Through the Lens of the Manifold Hypothesis
par: Peng, Xiong, et autres
Publié: (2025)

BadLabel: A Robust Perspective on Evaluating and Enhancing Label-noise Learning
par: Zhang, Jingfeng, et autres
Publié: (2023)

MeGU: Machine-Guided Unlearning with Target Feature Disentanglement
par: Wang, Haoyu, et autres
Publié: (2026)

Initialization Matters: On the Benign Overfitting of Two-Layer ReLU CNN with Fully Trainable Layers
par: Shang, Shuning, et autres
Publié: (2024)

Provable Robust Overfitting Mitigation in Wasserstein Distributionally Robust Optimization
par: Liu, Shuang, et autres
Publié: (2025)

The Surprising Harmfulness of Benign Overfitting for Adversarial Robustness
par: Hao, Yifan, et autres
Publié: (2024)

Where to Steer: Input-Dependent Layer Selection for Steering Improves LLM Alignment
par: Gadgil, Soham, et autres
Publié: (2026)

Catastrophic Overfitting, Entropy Gap and Participation Ratio: A Noiseless $l^p$ Norm Solution for Fast Adversarial Training
par: Mehouachi, Fares B., et autres
Publié: (2025)

Enhancing Certified Robustness via Block Reflector Orthogonal Layers and Logit Annealing Loss
par: Lai, Bo-Han, et autres
Publié: (2025)

Mitigating Structural Overfitting: A Distribution-Aware Rectification Framework for Missing Feature Imputation
par: Song, Yifan, et autres
Publié: (2025)

Auto-Prompt Generation is Not Robust: Prompt Optimization Driven by Pseudo Gradient
par: Shi, Zeru, et autres
Publié: (2024)

Testing for Overfitting
par: Schmidt, James
Publié: (2023)

What If the Input is Expanded in OOD Detection?
par: Zhang, Boxuan, et autres
Publié: (2024)

2-in-1 Accelerator: Enabling Random Precision Switch for Winning Both Adversarial Robustness and Efficiency
par: Fu, Yonggan, et autres
Publié: (2021)

Robust Policy Optimization to Prevent Catastrophic Forgetting
par: Sabbaghi, Mahdi, et autres
Publié: (2026)

Relative Overfitting and Accept-Reject Framework
par: Liu, Yanxin, et autres
Publié: (2025)

HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmark
par: Li, Chaojian, et autres
Publié: (2021)

Benign Overfitting for Regression with Trained Two-Layer ReLU Networks
par: Park, Junhyung, et autres
Publié: (2024)

Do LLMs Overcome Shortcut Learning? An Evaluation of Shortcut Challenges in Large Language Models
par: Yuan, Yu, et autres
Publié: (2024)