Differentiable Window for Dynamic Local Attention

by   Thanh-Tung Nguyen, et al.
Agency for Science, Technology and Research

We propose Differentiable Window, a new neural module and general purpose component for dynamic window selection. While universally applicable, we demonstrate a compelling use case of utilizing Differentiable Window to improve standard attention modules by enabling more focused attentions over the input regions. We propose two variants of Differentiable Window, and integrate them within the Transformer architecture in two novel ways. We evaluate our proposed approach on a myriad of NLP tasks, including machine translation, sentiment analysis, subject-verb agreement and language modeling. Our experimental results demonstrate consistent and sizable improvements across all tasks.


A differentiable short-time Fourier transform with respect to the window length

In this paper, we revisit the use of spectrograms in neural networks, by...

Memory Transformer

Transformer-based models have achieved state-of-the-art results in many ...

FastTrees: Parallel Latent Tree-Induction for Faster Sequence Encoding

Inducing latent tree structures from sequential data is an emerging tren...

Beyond Fixation: Dynamic Window Visual Transformer

Recently, a surge of interest in visual transformers is to reduce the co...

A Dynamic Window Neural Network for CCG Supertagging

Combinatory Category Grammar (CCG) supertagging is a task to assign lexi...

Masked Autoencoders with Multi-Window Attention Are Better Audio Learners

Several recent works have adapted Masked Autoencoders (MAEs) for learnin...

MuFuRU: The Multi-Function Recurrent Unit

Recurrent neural networks such as the GRU and LSTM found wide adoption i...

Please sign up or login with your details

Forgot password? Click here to reset