Yang Zhang (张阳)

I have published more than 50 peer-reviewed papers so far. My publication list can also be found at DBLP and Google Scholar, however, they may not be up to date. Note that in the domain of information security, the most prestigious conferences are IEEE S&P, CCS, USENIX Security, and NDSS. Also, the author list of most of my papers before 2018 follows alphabetical order.

2024

SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models

Boyang Zhang, Zheng Li, Ziqing Yang, Xinlei He, Michael Backes, Mario Fritz, Yang Zhang; USENIX Security 2024

2024

SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models

Quantifying Privacy Risks of Prompts in Visual Prompt Learning

You Only Prompt Once: On the Capabilities of Prompt Learning on Large Language Models to Tackle Toxic Content

Test-Time Poisoning Attacks Against Test-Time Adaptation Models

2023

DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models

Unsafe Diffusion: On the Generation of Unsafe Images and Hateful Memes From Text-To-Image Models

A Plot is Worth a Thousand Words: Model Information Stealing Attacks via Scientific Plots

Two-in-One: A Model Hijacking Attack Against Text Generation Models

UnGANable: Defending Against GAN-based Face Manipulation

FACE-AUDITOR: Data Auditing in Facial Recognition Systems

PrivTrace: Differentially Private Trajectory Synthesis by Adaptive Markov Model

Generated Graph Detection

Data Poisoning Attacks Against Multimodal Encoders

NOTABLE: Transferable Backdoor Attacks Against Prompt-based NLP Models

Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image Encoders

On the Evolution of (Hateful) Memes by Means of Multimodal Contrastive Learning

Is Adversarial Training Really a Silver Bullet for Mitigating Data Poisoning?

Backdoor Attacks Against Dataset Distillation

Pseudo Label-Guided Model Inversion Attack via Conditional Generative Adversarial Network

VERITRAIN: Validating MLaaS Training Efforts via Anomaly Detection

Robustness Over Time: Understanding Adversarial Examples' Effectiveness on Longitudinal Versions of Large Language Models

White-box Membership Inference Attacks against Diffusion Models

"Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models

Mondrian: Prompt Abstraction Attack Against Large Language Models for Cheaper API Pricing

Watermarking Diffusion Model

In ChatGPT We Trust? Measuring and Characterizing the Reliability of ChatGPT

MGTBench: Benchmarking Machine-Generated Text Detection

Prompt Stealing Attacks Against Text-to-Image Generation Models

Generative Watermarking Against Unauthorized Subject-Driven Image Synthesis

2022

Amplifying Membership Exposure via Data Poisoning

Why So Toxic? Measuring and Triggering Toxic Behavior in Open-Domain Chatbots

On the Privacy Risks of Cell-Based NAS Architectures

Membership Inference Attacks by Exploiting Loss Trajectory

Auditing Membership Leakages of Multi-Exit Networks

Graph Unlearning

SSLGuard: A Watermarking Scheme for Self-supervised Learning Pre-trained Encoders

Finding MNEMON: Reviving Memories of Node Embeddings

Semi-Leak: Membership Inference Attacks Against Semi-supervised Learning

Teacher Model Fingerprinting Attacks Against Transfer Learning

ML-Doctor: Holistic Risk Assessment of Inference Attacks Against Machine Learning Models

Inference Attacks Against Graph Neural Networks

On Xing Tian and the Perseverance of Anti-China Sentiment Online

Model Stealing Attacks Against Inductive Graph Neural Networks

Get a Model! Model Hijacking Attack Against Machine Learning Models

Property Inference Attacks Against GANs

Dynamic Backdoor Attacks Against Machine Learning Models

FairSR: Fairness-aware Sequential Recommendation through Multi-Task Learning with Preference Graph Embeddings

Fine-Tuning Is All You Need to Mitigate Backdoor Attacks

Membership Inference Attacks Against Text-to-image Generation Models

Backdoor Attacks in the Supply Chain of Masked Image Modeling

Membership-Doctor: Comprehensive Assessment of Membership Inference Against Machine Learning Models

2021

Quantifying and Mitigating Privacy Risks of Contrastive Learning

When Machine Unlearning Jeopardizes Privacy

Membership Inference Attacks Against Recommender Systems

Membership Leakage in Label-Only Exposures

BadNL: Backdoor Attacks Against NLP Models with Semantic-preserving Improvements

Stealing Links from Graph Neural Networks

PrivSyn: Differentially Private Data Synthesis

“Go eat a bat, Chang!”: On the Emergence of Sinophobic Behavior on Web Communities in the Face of COVID-19

DatingSec: Detecting Malicious Accounts in Dating Apps Using a Content-Based Attention Network

Node-Level Membership Inference Attacks Against Graph Neural Networks

2020

LEAF: A Faster Secure Search Algorithm via Localization, Extraction, and Reconstruction

GAN-Leaks: A Taxonomy of Membership Inference Attacks against Generative Models

Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning

Membership Inference Against DNA Methylation Databases

Towards Plausible Graph Anonymization

Don't Trigger Me! A Triggerless Backdoor Attack Against Deep Neural Networks

BAAAN: Backdoor Attacks Against Autoencoder and GAN-Based Machine Learning Models

Privacy Analysis of Deep Learning in the Wild: Membership Inference Attacks against Transfer Learning

2019

MemGuard: Defending against Black-Box Membership Inference Attacks via Adversarial Examples

How to Prove Your Model Belongs to You: A Blind-Watermark based Framework to Protect Intellectual Property of DNN

A Graph-Based Approach to Explore Relationship Between Hashtags and Images

Fairwalk: Towards Fair Graph Embedding

Language in Our Time: An Empirical Analysis of Hashtags