Image de couverture de livre

Subspace Defense: Discarding Adversarial Perturbations by Learning a Subspace for Clean Signals

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zheng, Rui, Zhou, Yuhao, Xi, Zhiheng, Gui, Tao, Zhang, Qi, Huang, Xuanjing
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning Computation and Language Cryptography and Security
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

CENSOR: Defense Against Gradient Inversion via Orthogonal Subspace Bayesian Sampling
par: Zhang, Kaiyuan, et autres
Publié: (2025)

Differentially Private Subspace Fine-Tuning for Large Language Models
par: Zheng, Lele, et autres
Publié: (2026)

FedBAP: Backdoor Defense via Benign Adversarial Perturbation in Federated Learning
par: Yan, Xinhai, et autres
Publié: (2025)

CEE: An Inference-Time Jailbreak Defense for Embodied Intelligence via Subspace Concept Rotation
par: Yang, Jirui, et autres
Publié: (2025)

Self-Evaluation as a Defense Against Adversarial Attacks on LLMs
par: Brown, Hannah, et autres
Publié: (2024)

Saliency Attention and Semantic Similarity-Driven Adversarial Perturbation
par: Waghela, Hetvi, et autres
Publié: (2024)

IDEA: Invariant Defense for Graph Adversarial Robustness
par: Tao, Shuchang, et autres
Publié: (2023)

CR-UTP: Certified Robustness against Universal Text Perturbations on Large Language Models
par: Lou, Qian, et autres
Publié: (2024)

AdvJudge-Zero: Binary Decision Flips in LLM-as-a-Judge via Adversarial Control Tokens
par: Li, Tung-Ling, et autres
Publié: (2025)

Probing Latent Subspaces in LLM for AI Security: Identifying and Manipulating Adversarial States
par: Chia, Xin Wei, et autres
Publié: (2025)

Promoting Data and Model Privacy in Federated Learning through Quantized LoRA
par: Zhu, JianHao, et autres
Publié: (2024)

AutoDefense: Multi-Agent LLM Defense against Jailbreak Attacks
par: Zeng, Yifan, et autres
Publié: (2024)

Hijacking Large Language Models via Adversarial In-Context Learning
par: Zhou, Xiangyu, et autres
Publié: (2023)

Adversarial Decoding: Generating Readable Documents for Adversarial Objectives
par: Zhang, Collin, et autres
Publié: (2024)

Low Rank Adaptation for Adversarial Perturbation
par: Liu, Han, et autres
Publié: (2026)

Combining Machine Learning Defenses without Conflicts
par: Duddu, Vasisht, et autres
Publié: (2024)

Adversarial Text Purification: A Large Language Model Approach for Defense
par: Moraffah, Raha, et autres
Publié: (2024)

PECAN: A Deterministic Certified Defense Against Backdoor Attacks
par: Zhang, Yuhao, et autres
Publié: (2023)

Mitigating the Backdoor Effect for Multi-Task Model Merging via Safety-Aware Subspace
par: Yang, Jinluan, et autres
Publié: (2024)

MIST: Defending Against Membership Inference Attacks Through Membership-Invariant Subspace Training
par: Li, Jiacheng, et autres
Publié: (2023)

MPAT: Building Robust Deep Neural Networks against Textual Adversarial Attacks
par: Zhang, Fangyuan, et autres
Publié: (2024)

Can Federated Learning Safeguard Private Data in LLM Training? Vulnerabilities, Attacks, and Defense Evaluation
par: Guo, Wenkai, et autres
Publié: (2025)

Testing the Limits of Jailbreaking Defenses with the Purple Problem
par: Kim, Taeyoun, et autres
Publié: (2024)

TwinGate: Stateful Defense against Decompositional Jailbreaks in Untraceable Traffic via Asymmetric Contrastive Learning
par: Sun, Bowen, et autres
Publié: (2026)

Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing
par: Xin, Yuan, et autres
Publié: (2023)

A Defensive Framework Against Adversarial Attacks on Machine Learning-Based Network Intrusion Detection Systems
par: Tafreshian, Benyamin, et autres
Publié: (2025)

From Theory to Practice: Evaluating Data Poisoning Attacks and Defenses in In-Context Learning on Social Media Health Discourse
par: Jhuma, Rabeya Amin, et autres
Publié: (2025)

Text-CRS: A Generalized Certified Robustness Framework against Textual Adversarial Attacks
par: Zhang, Xinyu, et autres
Publié: (2023)

Advancing Adversarial Suffix Transfer Learning on Aligned Large Language Models
par: Liu, Hongfu, et autres
Publié: (2024)

Humanizing Machine-Generated Content: Evading AI-Text Detection through Adversarial Attack
par: Zhou, Ying, et autres
Publié: (2024)

Data-centric NLP Backdoor Defense from the Lens of Memorization
par: Wang, Zhenting, et autres
Publié: (2024)

A No-Defense Defense Against Gradient-Based Adversarial Attacks on ML-NIDS: Is Less More?
par: elShehaby, Mohamed, et autres
Publié: (2026)

Adversarial Suffix Filtering: a Defense Pipeline for LLMs
par: Khachaturov, David, et autres
Publié: (2025)

Enhancing the "Immunity" of Mixture-of-Experts Networks for Adversarial Defense
par: Han, Qiao, et autres
Publié: (2024)

CLIBE: Detecting Dynamic Backdoors in Transformer-based NLP Models
par: Zeng, Rui, et autres
Publié: (2024)

How Vulnerable Is My Learned Policy? Universal Adversarial Perturbation Attacks On Modern Behavior Cloning Policies
par: Kalra, Akansha, et autres
Publié: (2025)

PromptRobust: Towards Evaluating the Robustness of Large Language Models on Adversarial Prompts
par: Zhu, Kaijie, et autres
Publié: (2023)

Test-time Adversarial Defense with Opposite Adversarial Path and High Attack Time Cost
par: Yeh, Cheng-Han, et autres
Publié: (2024)

One Stone, Two Birds: Enhancing Adversarial Defense Through the Lens of Distributional Discrepancy
par: Zhang, Jiacheng, et autres
Publié: (2025)

Break the Breakout: Reinventing LM Defense Against Jailbreak Attacks with Self-Refinement
par: Kim, Heegyu, et autres
Publié: (2024)