EchoMask employs reconfigurable acoustic metamaterials to generate physical-layer frequency-selective interference that raises voiceprint miss-match rates above 90% across tested microphones while preserving speech intelligibility.
Privacy leakage on dnns: A survey of model inversion attacks and defenses
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 5roles
background 1polarities
background 1representative citing papers
World models enable efficient AI planning but create risks from adversarial corruption, goal misgeneralization, and human bias, demonstrated via attacks that amplify errors and reduce rewards on models like RSSM and DreamerV3.
CatShift detects training data membership in LLMs by comparing output shifts induced by fine-tuning on member versus non-member data, relying on catastrophic forgetting without requiring logit access.
Gradient leakage attacks on GNNs for netlist benchmarks can expose gate types and Trojan properties; attention-based models leak more while defenses like differential privacy help only in limited cases without full performance preservation.
TS-LFO is a two-stage latent feature optimization method that bypasses state-of-the-art copyright defenses in diffusion-based image customization by restoring semantic consistency in latent space.
citing papers explorer
-
Before the Mic: Physical-Layer Voiceprint Anonymization with Acoustic Metamaterials
EchoMask employs reconfigurable acoustic metamaterials to generate physical-layer frequency-selective interference that raises voiceprint miss-match rates above 90% across tested microphones while preserving speech intelligibility.
-
Safety, Security, and Cognitive Risks in World Models
World models enable efficient AI planning but create risks from adversarial corruption, goal misgeneralization, and human bias, demonstrated via attacks that amplify errors and reduce rewards on models like RSSM and DreamerV3.
-
Hey, That's My Data! Token-Only Dataset Inference in Large Language Models
CatShift detects training data membership in LLMs by comparing output shifts induced by fine-tuning on member versus non-member data, relying on catastrophic forgetting without requiring logit access.
-
Leaking Circuit Secrets: Gradient Leakage Attacks on Graph Neural Networks
Gradient leakage attacks on GNNs for netlist benchmarks can expose gate types and Trojan properties; attention-based models leak more while defenses like differential privacy help only in limited cases without full performance preservation.
-
Bypassing Copyright Protection in Diffusion-based Customization via Two-Stage Latent Feature Optimization
TS-LFO is a two-stage latent feature optimization method that bypasses state-of-the-art copyright defenses in diffusion-based image customization by restoring semantic consistency in latent space.