Picture for Shaofeng Li

Shaofeng Li

Intriguing Frequency Interpretation of Adversarial Robustness for CNNs and ViTs

Add code
Jun 15, 2025
Viaarxiv icon

Model Inversion in Split Learning for Personalized LLMs: New Insights from Information Bottleneck Theory

Add code
Jan 10, 2025
Viaarxiv icon

FD2-Net: Frequency-Driven Feature Decomposition Network for Infrared-Visible Object Detection

Add code
Dec 12, 2024
Figure 1 for FD2-Net: Frequency-Driven Feature Decomposition Network for Infrared-Visible Object Detection
Figure 2 for FD2-Net: Frequency-Driven Feature Decomposition Network for Infrared-Visible Object Detection
Figure 3 for FD2-Net: Frequency-Driven Feature Decomposition Network for Infrared-Visible Object Detection
Figure 4 for FD2-Net: Frequency-Driven Feature Decomposition Network for Infrared-Visible Object Detection
Viaarxiv icon

Show Me What and Where has Changed? Question Answering and Grounding for Remote Sensing Change Detection

Add code
Oct 31, 2024
Figure 1 for Show Me What and Where has Changed? Question Answering and Grounding for Remote Sensing Change Detection
Figure 2 for Show Me What and Where has Changed? Question Answering and Grounding for Remote Sensing Change Detection
Figure 3 for Show Me What and Where has Changed? Question Answering and Grounding for Remote Sensing Change Detection
Figure 4 for Show Me What and Where has Changed? Question Answering and Grounding for Remote Sensing Change Detection
Viaarxiv icon

Unbridled Icarus: A Survey of the Potential Perils of Image Inputs in Multimodal Large Language Model Security

Add code
Apr 08, 2024
Viaarxiv icon

Seeing is not always believing: The Space of Harmless Perturbations

Add code
Feb 03, 2024
Figure 1 for Seeing is not always believing: The Space of Harmless Perturbations
Figure 2 for Seeing is not always believing: The Space of Harmless Perturbations
Figure 3 for Seeing is not always believing: The Space of Harmless Perturbations
Figure 4 for Seeing is not always believing: The Space of Harmless Perturbations
Viaarxiv icon

Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations

Add code
Feb 22, 2022
Figure 1 for Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations
Figure 2 for Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations
Figure 3 for Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations
Figure 4 for Fingerprinting Deep Neural Networks Globally via Universal Adversarial Perturbations
Viaarxiv icon

Exposing Weaknesses of Malware Detectors with Explainability-Guided Evasion Attacks

Add code
Nov 19, 2021
Figure 1 for Exposing Weaknesses of Malware Detectors with Explainability-Guided Evasion Attacks
Figure 2 for Exposing Weaknesses of Malware Detectors with Explainability-Guided Evasion Attacks
Figure 3 for Exposing Weaknesses of Malware Detectors with Explainability-Guided Evasion Attacks
Figure 4 for Exposing Weaknesses of Malware Detectors with Explainability-Guided Evasion Attacks
Viaarxiv icon

Hidden Backdoors in Human-Centric Language Models

Add code
May 01, 2021
Figure 1 for Hidden Backdoors in Human-Centric Language Models
Figure 2 for Hidden Backdoors in Human-Centric Language Models
Figure 3 for Hidden Backdoors in Human-Centric Language Models
Figure 4 for Hidden Backdoors in Human-Centric Language Models
Viaarxiv icon

Deep Learning Backdoors

Add code
Jul 16, 2020
Figure 1 for Deep Learning Backdoors
Figure 2 for Deep Learning Backdoors
Figure 3 for Deep Learning Backdoors
Figure 4 for Deep Learning Backdoors
Viaarxiv icon