Xingjun Ma

26

Papers

77

Total Citations

1

Affiliations

Affiliations

Fudan University

Papers (26)

Adversarial Prompt Tuning for Vision-Language Models

BlueSuffix: Reinforced Blue Teaming for Vision-Language Models Against Jailbreak Attacks

LDReg: Local Dimensionality Regularized Self-Supervised Learning

Anyattack: Towards Large-scale Self-supervised Adversarial Attacks on Vision-language Models

Free-Form Motion Control: Controlling the 6D Poses of Camera and Objects in Video Generation

AIM: Additional Image Guided Generation of Transferable Adversarial Attacks

HoneypotNet: Backdoor Attacks Against Model Extraction

Unlearnable Clusters: Towards Label-Agnostic Unlearnable Examples

Symmetric Cross Entropy for Robust Learning With Noisy Labels

Revisiting Adversarial Robustness Distillation: Robust Soft Labels Make Student Better

Short-Term and Long-Term Context Aggregation Network for Video Inpainting

Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks

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

Towards Million-Scale Adversarial Robustness Evaluation With Stronger Individual Attacks

StolenLoRA: Exploring LoRA Extraction Attacks via Synthetic Data

IDEATOR: Jailbreaking and Benchmarking Large Vision-Language Models Using Themselves

Iterative Learning With Open-Set Noisy Labels

Adversarial Camouflage: Hiding Physical-World Attacks With Natural Styles

Clean-Label Backdoor Attacks on Video Recognition Models

Exploring Architectural Ingredients of Adversarially Robust Deep Neural Networks

Anti-Backdoor Learning: Training Clean Models on Poisoned Data

Gradient Driven Rewards to Guarantee Fairness in Collaborative Machine Learning

$\alpha$-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression

CalFAT: Calibrated Federated Adversarial Training with Label Skewness

Dimensionality-Driven Learning with Noisy Labels

On the Convergence and Robustness of Adversarial Training