A Practical Survey on Faster and Lighter Transformers

By Quentin Fournier, Gaétan Marceau Caron, Daniel J. Aloise. 2021

167 citations recorded by the source corpus.

Recurrent neural networks are effective models to process sequences. However, they are unable to learn long-term dependencies because of their inherent sequential nature. As a solution, Vaswani et al. introduced the Transformer, a model solely based on the attention mechanism that is able to relate any two positions of the input sequence, hence modelling arbitrary long dependencies. The Transformer has improved the state-of-the-art across numerous sequence modelling tasks. However, its effectiveness comes at the expense of a quadratic computational and memory complexity with respect to the sequence length, hindering its adoption. Fortunately, the deep learning community has always been interested in improving the models’ efficiency, leading to a plethora of solutions such as parameter sharing, pruning, mixed-precision, and knowledge distillation. Recently, researchers have directly addressed the Transformer’s limitation by designing lower-complexity alternatives such as the Longformer, Reformer, Linformer, and Performer. However, due to the wide range of solutions, it has become challenging for researchers and practitioners to determine which methods to apply in practice to meet the desired tradeoff between capacity, computation, and memory. This survey addresses this issue by investigating popular approaches to make Transformers faster and lighter and by providing a comprehensive explanation of the methods’ strengths, limitations, and underlying assumptions.

Subfields: Computer Networks and Communications, Information Systems, Materials Chemistry, Model Architectures, Efficiency & Compression

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Information Systems

In Materials Chemistry

In Model Architectures

In Efficiency & Compression

Citations in the corpus

SCROLLS: Standardized CompaRison Over Long Language Sequences (2022, background): “To address this problem, a wide variety of efﬁcient alternatives and approximations have been proposed over the past couple of years (Tay et al., 2020b; Fournier et al., 2021).”
HyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution (2023, methodology): “Previous methods to reduce the quadratic cost of attention have used specialized methods to approximate full dense attention [18].”