Adversarial Perturbation

Adversarial perturbation research focuses on developing and mitigating the vulnerability of machine learning models to maliciously crafted inputs designed to cause misclassification or other errors. Current research emphasizes improving the robustness of various model architectures, including deep convolutional neural networks, vision transformers, and graph neural networks, often employing techniques like adversarial training, vector quantization, and optimal transport methods. This field is crucial for ensuring the reliability and security of AI systems across diverse applications, from image classification and face recognition to robotics and natural language processing, by identifying and addressing vulnerabilities to attacks.

423papers

Papers - Page 4

November 20, 2024

TAPT: Test-Time Adversarial Prompt Tuning for Robust Inference in Vision-Language Models
Xin Wang, Kai Chen, Jiaming Zhang, Jingjing Chen, Xingjun Ma
Vision Language Model Robust Inference Zero Shot Adversarial Perturbation

November 18, 2024

CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization
Nay Myat Min, Long H. Pham, Yige Li, Jun Sun
Consistency Regularization Adversarial Perturbation Backdoor Attack Backdoor Defense

October 31, 2024

Wide Two-Layer Networks can Learn from Adversarial Perturbations
Soichiro Kumano, Hiroshi Kera, Toshihiko Yamasaki
Class Specific Feature Adversarial Example Two Layer Network Adversarial Perturbation

October 30, 2024

October 29, 2024

CausAdv: A Causal-based Framework for Detecting Adversarial Examples
Hichem Debbi
Adversarial Perturbation Adversarial Text Adversarial Robustness Adversarial Example Causal Feature Causal Approach

October 26, 2024

Attacks against Abstractive Text Summarization Models through Lead Bias and Influence Functions
Poojitha Thota, Shirin Nilizadeh
Influence Function New Attack Adversarial Perturbation Abstractive Summary Text Summary Systematic Bias Abstractive Summarization Model Adversarial Robustness

October 21, 2024

October 20, 2024

PEAS: A Strategy for Crafting Transferable Adversarial Examples
Bar Avraham, Yisroel Mirsky
Adversarial Perturbation Adversary Agent Supervised ImageNet Adversarial Example Strategy Logic

October 17, 2024

October 16, 2024

October 15, 2024

October 14, 2024

Towards Calibrated Losses for Adversarial Robust Reject Option Classification
Vrund Shah, Tejas Chaudhari, Naresh Manwani
Adversarial Attack Adversarial Perturbation Classification Code Adversarial Setting

October 12, 2024

Impeding LLM-assisted Cheating in Introductory Programming Assignments via Adversarial Perturbation
Saiful Islam Salim, Rubin Yuchan Yang, Alexander Cooper, Suryashree Ray, Saumya Debray, Sazzadur Rahaman
Introductory Programming Code Generation Adversarial Perturbation Cheating Detection

October 11, 2024

Adversarial Training Can Provably Improve Robustness: Theoretical Analysis of Feature Learning Process Under Structured Data
Binghui Li, Yuanzhi Li
Structured Data Native Robustness Model Robustness Feature Learning Adversarial Perturbation Adversarial Training Adversarial Example General Analysis

Adversarial Perturbation

Papers - Page 4

TAPT: Test-Time Adversarial Prompt Tuning for Robust Inference in Vision-Language Models

CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization

Wide Two-Layer Networks can Learn from Adversarial Perturbations

Keep on Swimming: Real Attackers Only Need Partial Knowledge of a Multi-Model System

Teaching a Language Model to Distinguish Between Similar Details using a Small Adversarial Training Set

CausAdv: A Causal-based Framework for Detecting Adversarial Examples

Attacks against Abstractive Text Summarization Models through Lead Bias and Influence Functions

Robust Feature Learning for Multi-Index Models in High Dimensions

Model Mimic Attack: Knowledge Distillation for Provably Transferable Adversarial Examples

PEAS: A Strategy for Crafting Transferable Adversarial Examples

MMAD-Purify: A Precision-Optimized Framework for Efficient and Scalable Multi-Modal Attacks

Boosting Imperceptibility of Stable Diffusion-based Adversarial Examples Generation with Momentum

Perseus: Leveraging Common Data Patterns with Curriculum Learning for More Robust Graph Neural Networks

DAT: Improving Adversarial Robustness via Generative Amplitude Mix-up in Frequency Domain

Efficient and Effective Universal Adversarial Attack against Vision-Language Pre-training Models

Automatically Generating Visual Hallucination Test Cases for Multimodal Large Language Models

Adversarially Guided Stateful Defense Against Backdoor Attacks in Federated Deep Learning

Towards Calibrated Losses for Adversarial Robust Reject Option Classification

Impeding LLM-assisted Cheating in Introductory Programming Assignments via Adversarial Perturbation

Adversarial Training Can Provably Improve Robustness: Theoretical Analysis of Feature Learning Process Under Structured Data