Abstract: Automatic speech recognition (ASR) models are frequently exposed to data distribution shifts in many real-world scenarios, leading to erroneous predictions. To tackle this issue, an existing ...

But having tested this setting on various vacuum cleaners, I came to realize that not all ‘auto’ modes are created equal.

Abstract: Detecting advanced persistent threats (APTs) at a host via data provenance has emerged as a valuable yet challenging task. Compared with attack rule matching, machine learning approaches ...