https://arxiv.org/abs/2110.13711

close this message

Thank you for supporting arXiv

Thank you to everyone who donated during arXiv's Giving Week, October
25 - 31. It's not too late to give. arXiv is a nonprofit that depends
on donations to fund essential operations and new initiatives. We
appreciate your support of Open Science. Thank you!

DONATE

[secure site, no need to create account]

Skip to main content
Cornell University
We gratefully acknowledge support from
the Simons Foundation and member institutions.
 
arXiv.org > cs > arXiv:2110.13711
[                    ]

Help | Advanced Search

[All fields        ]
Search
arXiv
Cornell University Logo
[                    ] GO
quick links

  * Login
  * Help Pages
  * About

Computer Science > Machine Learning

arXiv:2110.13711 (cs)
[Submitted on 26 Oct 2021]

Title:Hierarchical Transformers Are More Efficient Language Models

Authors:Piotr Nawrot, Szymon Tworkowski, Michal Tyrolski, Lukasz
Kaiser, Yuhuai Wu, Christian Szegedy, Henryk Michalewski
Download PDF

    Abstract: Transformer models yield impressive results on many NLP
    and sequence modeling tasks. Remarkably, Transformers can handle
    long sequences which allows them to produce long coherent
    outputs: full paragraphs produced by GPT-3 or well-structured
    images produced by DALL-E. These large language models are
    impressive but also very inefficient and costly, which limits
    their applications and accessibility. We postulate that having an
    explicit hierarchical architecture is the key to Transformers
    that efficiently handle long sequences. To verify this claim, we
    first study different ways to downsample and upsample activations
    in Transformers so as to make them hierarchical. We use the best
    performing upsampling and downsampling layers to create Hourglass
    - a hierarchical Transformer language model. Hourglass improves
    upon the Transformer baseline given the same amount of
    computation and can yield the same results as Transformers more
    efficiently. In particular, Hourglass sets new state-of-the-art
    for Transformer models on the ImageNet32 generation task and
    improves language modeling efficiency on the widely studied
    enwik8 benchmark.

Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Cite as:  arXiv:2110.13711 [cs.LG]
          (or arXiv:2110.13711v1 [cs.LG] for this version)

Submission history

From: Szymon Tworkowski [view email]
[v1] Tue, 26 Oct 2021 14:00:49 UTC (1,841 KB)
Full-text links:

Download:

  * PDF
  * Other formats

(license)
Current browse context:
cs.LG
< prev   |   next >
new | recent | 2110
Change to browse by:
cs
cs.CL

References & Citations

  * NASA ADS
  * Google Scholar
  * Semantic Scholar

a export bibtex citation Loading...

Bibtex formatted citation

x
[loading...          ]
Data provided by:

Bookmark

BibSonomy logo Mendeley logo Reddit logo ScienceWISE logo
(*) Bibliographic Tools

Bibliographic and Citation Tools

[ ] Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
[ ] Litmaps Toggle
Litmaps (What is Litmaps?)
[ ] scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
( ) Code & Data

Code and Data Associated with this Article

[ ] arXiv Links to Code Toggle
arXiv Links to Code & Data (What is Links to Code & Data?)
( ) Related Papers

Recommenders and Search Tools

[ ] Connected Papers Toggle
Connected Papers (What is Connected Papers?)
[ ] Core recommender toggle
CORE Recommender (What is CORE?)
( ) About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and
share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have
embraced and accepted our values of openness, community, excellence,
and user data privacy. arXiv is committed to these values and only
works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community?
Learn more about arXivLabs and how to get involved.

Which authors of this paper are endorsers? | Disable MathJax (What is
MathJax?)

  * About
  * Help

  * Click here to contact arXiv Contact
  * Click here to subscribe Subscribe

  * Copyright
  * Privacy Policy

  * Web Accessibility Assistance
  * arXiv Operational Status
    Get status notifications via email or slack