Picture for Leo Yu Zhang

Leo Yu Zhang

Memorization in deep learning: A survey

Jun 06, 2024
Viaarxiv icon

Large Language Model Watermark Stealing With Mixed Integer Programming

May 30, 2024
Viaarxiv icon

IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency

Add code
May 16, 2024
Figure 1 for IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency
Figure 2 for IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency
Figure 3 for IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency
Figure 4 for IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency
Viaarxiv icon

Algorithmic Fairness: A Tolerance Perspective

Apr 26, 2024
Viaarxiv icon

Detector Collapse: Backdooring Object Detection to Catastrophic Overload or Blindness

Add code
Apr 17, 2024
Viaarxiv icon

Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples

Add code
Mar 19, 2024
Figure 1 for Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples
Figure 2 for Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples
Figure 3 for Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples
Figure 4 for Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples
Viaarxiv icon

Fluent: Round-efficient Secure Aggregation for Private Federated Learning

Add code
Mar 10, 2024
Figure 1 for Fluent: Round-efficient Secure Aggregation for Private Federated Learning
Figure 2 for Fluent: Round-efficient Secure Aggregation for Private Federated Learning
Figure 3 for Fluent: Round-efficient Secure Aggregation for Private Federated Learning
Figure 4 for Fluent: Round-efficient Secure Aggregation for Private Federated Learning
Viaarxiv icon

Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks

Jan 30, 2024
Viaarxiv icon

MISA: Unveiling the Vulnerabilities in Split Federated Learning

Dec 19, 2023
Viaarxiv icon

Corrupting Convolution-based Unlearnable Datasets with Pixel-based Image Transformations

Add code
Nov 30, 2023
Viaarxiv icon