# FNetAR: Mixing Tokens with Autoregressive Fourier Transforms

arXiv: https://arxiv.org/abs/2107.10932 · HTML: https://mlsanity.com/paper/2107.10932/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 15 references that are in the map.
- First posted on arXiv 22 Jul 2021.

## Builds on (most cited first, 11 of 11)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://mlsanity.com/paper/1906.08237/) (2019)
- [Non-local Neural Networks](https://mlsanity.com/paper/1711.07971/) (2017)
- [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://mlsanity.com/paper/1901.02860/) (2019)
- [Rethinking Attention with Performers](https://mlsanity.com/paper/2009.14794/) (2020)
- [FNet: Mixing Tokens with Fourier Transforms](https://mlsanity.com/paper/2105.03824/) (2021)
- [Masked Language Modeling for Proteins via Linearly Scalable Long-Context Transformers](https://mlsanity.com/paper/2006.03555/) (2020)
- [Controlling Computation versus Quality for Neural Sequence Models](https://mlsanity.com/paper/2002.07106/) (2020)
- [Fourier Image Transformer](https://mlsanity.com/paper/2104.02555/) (2021)

## Built on by (most cited first, 1 of 1)

- [Memory-efficient Stochastic methods for Memory-based Transformers](https://mlsanity.com/paper/2311.08123/) (2023)

## Abstract (arXiv)

In this note we examine the autoregressive generalization of the FNet algorithm, in which self-attention layers from the standard Transformer architecture are substituted with a trivial sparse-uniformsampling procedure based on Fourier transforms. Using the Wikitext-103 benchmark, we demonstratethat FNetAR retains state-of-the-art performance (25.8 ppl) on the task of causal language modelingcompared to a Transformer-XL baseline (24.2 ppl) with only half the number self-attention layers,thus providing further evidence for the superfluity of deep neural networks with heavily compoundedattention mechanisms. The autoregressive Fourier transform could likely be used for parameterreduction on most Transformer-based time-series prediction models.