Picture for Qiben Yan

Qiben Yan

Michigan State University

Privacy-Preserving Diffusion Model Using Homomorphic Encryption

Add code
Mar 09, 2024
Figure 1 for Privacy-Preserving Diffusion Model Using Homomorphic Encryption
Figure 2 for Privacy-Preserving Diffusion Model Using Homomorphic Encryption
Figure 3 for Privacy-Preserving Diffusion Model Using Homomorphic Encryption
Figure 4 for Privacy-Preserving Diffusion Model Using Homomorphic Encryption
Viaarxiv icon

A Practical Survey on Emerging Threats from AI-driven Voice Attacks: How Vulnerable are Commercial Voice Control Systems?

Dec 10, 2023
Viaarxiv icon

Beyond Boundaries: A Comprehensive Survey of Transferable Attacks on AI Systems

Nov 20, 2023
Viaarxiv icon

MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems

Add code
Sep 13, 2023
Figure 1 for MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems
Figure 2 for MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems
Figure 3 for MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems
Figure 4 for MASTERKEY: Practical Backdoor Attack Against Speaker Verification Systems
Viaarxiv icon

PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection

Sep 13, 2023
Figure 1 for PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection
Figure 2 for PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection
Figure 3 for PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection
Figure 4 for PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection
Viaarxiv icon

Understanding Multi-Turn Toxic Behaviors in Open-Domain Chatbots

Jul 14, 2023
Figure 1 for Understanding Multi-Turn Toxic Behaviors in Open-Domain Chatbots
Figure 2 for Understanding Multi-Turn Toxic Behaviors in Open-Domain Chatbots
Figure 3 for Understanding Multi-Turn Toxic Behaviors in Open-Domain Chatbots
Figure 4 for Understanding Multi-Turn Toxic Behaviors in Open-Domain Chatbots
Viaarxiv icon

VSMask: Defending Against Voice Synthesis Attack via Real-Time Predictive Perturbation

May 09, 2023
Figure 1 for VSMask: Defending Against Voice Synthesis Attack via Real-Time Predictive Perturbation
Figure 2 for VSMask: Defending Against Voice Synthesis Attack via Real-Time Predictive Perturbation
Figure 3 for VSMask: Defending Against Voice Synthesis Attack via Real-Time Predictive Perturbation
Figure 4 for VSMask: Defending Against Voice Synthesis Attack via Real-Time Predictive Perturbation
Viaarxiv icon

A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT

Add code
Feb 18, 2023
Figure 1 for A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT
Figure 2 for A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT
Figure 3 for A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT
Figure 4 for A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT
Viaarxiv icon

NEC: Speaker Selective Cancellation via Neural Enhanced Ultrasound Shadowing

Add code
Jul 12, 2022
Figure 1 for NEC: Speaker Selective Cancellation via Neural Enhanced Ultrasound Shadowing
Figure 2 for NEC: Speaker Selective Cancellation via Neural Enhanced Ultrasound Shadowing
Figure 3 for NEC: Speaker Selective Cancellation via Neural Enhanced Ultrasound Shadowing
Figure 4 for NEC: Speaker Selective Cancellation via Neural Enhanced Ultrasound Shadowing
Viaarxiv icon

SuperVoice: Text-Independent Speaker Verification Using Ultrasound Energy in Human Speech

Add code
May 28, 2022
Figure 1 for SuperVoice: Text-Independent Speaker Verification Using Ultrasound Energy in Human Speech
Figure 2 for SuperVoice: Text-Independent Speaker Verification Using Ultrasound Energy in Human Speech
Figure 3 for SuperVoice: Text-Independent Speaker Verification Using Ultrasound Energy in Human Speech
Figure 4 for SuperVoice: Text-Independent Speaker Verification Using Ultrasound Energy in Human Speech
Viaarxiv icon