NLP

Staircase Attention for Recurrent Processing of Sequences

December 29, 2022

Abstract

Attention mechanisms have become a standard tool for sequence modeling tasks, in particular by stacking self-attention layers over the entire input sequence as in the Transformer architecture. In this work we introduce a novel attention procedure called staircase attention that, unlike self-attention, operates across the sequence (in time) recurrently processing the input by adding another step of processing. A step in the staircase comprises of backward tokens (encoding the sequence so far seen) and forward tokens (ingesting a new part of the sequence). Thus our model can trade off performance and compute, by increasing the amount of recurrence through time and depth. Staircase attention is shown to be able to solve tasks that involve tracking that conventional Transformers cannot, due to this recurrence. Further, it is shown to provide improved modeling power for the same size model (number of parameters) compared to self-attentive Transformers on large language modeling and dialogue tasks, yielding significant perplexity gains.

Download the Paper

AUTHORS

Publisher

neurips

Related Publications

June 25, 2024

NLP

Neurons in Large Language Models: Dead, N-gram, Positional

Elena Voita, Javier Ferrando Monsonis, Christoforos Nalmpantis

June 25, 2024

June 25, 2024

SPEECH & AUDIO

NLP

Textless Acoustic Model with Self-Supervised Distillation for Noise-Robust Expressive Speech-to-Speech Translation

Min-Jae Hwang, Ilia Kulikov, Benjamin Peloquin, Hongyu Gong, Peng-Jen Chen, Ann Lee

June 25, 2024

June 14, 2024

NLP

How to Train Your DRAGON: Diverse Augmentation Towards Generalizable Dense Retrieval

Sheng-Chieh Lin, Akari Asai, Minghan Li, Barlas Oguz, Jimmy Lin, Scott Yih, Xilun Chen

June 14, 2024

June 14, 2024

NLP

SYSTEMS RESEARCH

LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding

Mostafa Elhoushi, Akshat Shrivastava, Diana Liskovich, Basil Hosmer, Bram Wasti, Liangzhen Lai, Nas Mahmoud, Bilge Acun, Saurabh Agarwal, Ahmed Roman, Ahmed Aly, Beidi Chen, Carole-Jean Wu

June 14, 2024

Help Us Pioneer The Future of AI

We share our open source frameworks, tools, libraries, and models for everything from research exploration to large-scale production deployment.