Gmail’s New Look Is About To Become Opt-Out Instead of Opt-In Extracting Targeted Training Data from ASR Models, and How to Mitigate It. (arXiv:2204.08345v2 [cs.SD] UPDATED)
Recent work has designed methods to demonstrate that model updates in ASR training can leak potentially sensitive attributes of the...
Improved Certified Defenses against Data Poisoning with (Deterministic) Finite Aggregation. (arXiv:2202.02628v2 [cs.LG] UPDATED)
Data poisoning attacks aim at manipulating model behaviors through distorting training data. Previously, an aggregation-based certified defense, Deep Partition Aggregation...
Detecting Unintended Memorization in Language-Model-Fused ASR. (arXiv:2204.09606v2 [cs.CL] UPDATED)
End-to-end (E2E) models are often being accompanied by language models (LMs) via shallow fusion for boosting their overall quality as...
Extracting Targeted Training Data from ASR Models, and How to Mitigate It. (arXiv:2204.08345v2 [cs.SD] UPDATED)
Recent work has designed methods to demonstrate that model updates in ASR training can leak potentially sensitive attributes of the...
Detecting Unintended Memorization in Language-Model-Fused ASR. (arXiv:2204.09606v2 [cs.CL] UPDATED)
End-to-end (E2E) models are often being accompanied by language models (LMs) via shallow fusion for boosting their overall quality as...
Multi-relational Instruction Association Graph for Cross-architecture Binary Similarity Comparison. (arXiv:2206.12236v2 [cs.CR] UPDATED)
Cross-architecture binary similarity comparison is essential in many security applications. Recently, researchers have proposed learning-based approaches to improve comparison performance....
Multi-relational Instruction Association Graph for Cross-architecture Binary Similarity Comparison. (arXiv:2206.12236v2 [cs.CR] UPDATED)
Cross-architecture binary similarity comparison is essential in many security applications. Recently, researchers have proposed learning-based approaches to improve comparison performance....
Debiasing Learning for Membership Inference Attacks Against Recommender Systems. (arXiv:2206.12401v2 [cs.IR] UPDATED)
Learned recommender systems may inadvertently leak information about their training data, leading to privacy violations. We investigate privacy threats faced...
Compose Your Own ’80s Tunes With These Super-Portable Synths
These tiny, adorable, rechargeable, and affordable synths fit in a backpack and make you sound like a modern-day A-ha.
Reverse Engineering An Apollo-Era Module with X-Ray
The gear that helped us walk on the Moon nearly 60 years ago is still giving up its mysteries today,...