Recurrent neural networks (RNNs) sequentially process data by updating their state with
each new data point, and have long been the de facto choice for sequence modeling tasks …

We present AdaFrame, a framework that adaptively selects relevant frames on a per-input
basis for fast video recognition. AdaFrame contains a Long Short-Term Memory network …

Deep learning has recently demonstrated state-of-the art performance on key tasks related
to the maintenance of computer systems, such as intrusion detection, denial of service attack …

The rapid growth of user-generated videos on the Internet has intensified the need for text-
based video retrieval systems. Traditional methods mainly favor the concept-based …

Abstract Language exhibits hierarchical structure, but recent work using a subject-verb
agreement diagnostic argued that state-of-the-art language models, LSTMs, fail to learn long …

Abstract We introduce Transformer Grammars (TGs), a novel class of Transformer language
models that combine (i) the expressive power, scalability, and strong performance of …

Both real and fake news in various domains, such as politics, health, and entertainment are
spread via online social media every day, necessitating fake news detection for multiple …

We introduce AdaFrame, a conditional computation framework that adaptively selects
relevant frames on a per-input basis for fast video recognition. AdaFrame, which contains a …

We introduce a neural stack architecture with a differentiable parameterized stack operator
approximating stack push and pop operations. We prove the stability of this stack …

For the security defense in the current Intelligent Transportation System (ITS), malware is
often used as the security analysis data source, but only the known attack type can be …