https://arxiv.org/abs/2404.07143

Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member
institutions, and all contributors. Donate
 
arxiv logo > cs > arXiv:2404.07143
[                    ]

Help | Advanced Search

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

  * Login
  * Help Pages
  * About

Computer Science > Computation and Language

arXiv:2404.07143 (cs)
[Submitted on 10 Apr 2024]

Title:Leave No Context Behind: Efficient Infinite Context
Transformers with Infini-attention

Authors:Tsendsuren Munkhdalai, Manaal Faruqui, Siddharth Gopal
View a PDF of the paper titled Leave No Context Behind: Efficient
Infinite Context Transformers with Infini-attention, by Tsendsuren
Munkhdalai and 1 other authors
View PDF HTML (experimental)

    Abstract:This work introduces an efficient method to scale
    Transformer-based Large Language Models (LLMs) to infinitely long
    inputs with bounded memory and computation. A key component in
    our proposed approach is a new attention technique dubbed
    Infini-attention. The Infini-attention incorporates a compressive
    memory into the vanilla attention mechanism and builds in both
    masked local attention and long-term linear attention mechanisms
    in a single Transformer block. We demonstrate the effectiveness
    of our approach on long-context language modeling benchmarks, 1M
    sequence length passkey context block retrieval and 500K length
    book summarization tasks with 1B and 8B LLMs. Our approach
    introduces minimal bounded memory parameters and enables fast
    streaming inference for LLMs.

Comments: 9 pages, 4 figures, 4 tables
          Computation and Language (cs.CL); Artificial Intelligence
Subjects: (cs.AI); Machine Learning (cs.LG); Neural and Evolutionary
          Computing (cs.NE)
Cite as:  arXiv:2404.07143 [cs.CL]
          (or arXiv:2404.07143v1 [cs.CL] for this version)
          https://doi.org/10.48550/arXiv.2404.07143
          Focus to learn more
          arXiv-issued DOI via DataCite

Submission history

From: Tsendsuren Munkhdalai [view email]
[v1] Wed, 10 Apr 2024 16:18:42 UTC (248 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Leave No Context Behind: Efficient
    Infinite Context Transformers with Infini-attention, by
    Tsendsuren Munkhdalai and 1 other authors
  * View PDF
  * HTML (experimental)
  * TeX Source
  * Other Formats

view license
Current browse context:
cs.CL
< prev   |   next >
new | recent | 2404
Change to browse by:
cs
cs.AI
cs.LG
cs.NE

References & Citations

  * NASA ADS
  * Google Scholar
  * Semantic Scholar

a export BibTeX citation Loading...

BibTeX formatted citation

x
[loading...          ]
Data provided by:

Bookmark

BibSonomy logo Reddit 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, Media

Code, Data and Media Associated with this Article

[ ] Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
[ ] DagsHub Toggle
DagsHub (What is DagsHub?)
[ ] GotitPub Toggle
Gotit.pub (What is GotitPub?)
[ ] Links to Code Toggle
Papers with Code (What is Papers with Code?)
[ ] ScienceCast Toggle
ScienceCast (What is ScienceCast?)
( ) Demos

Demos

[ ] Replicate Toggle
Replicate (What is Replicate?)
[ ] Spaces Toggle
Hugging Face Spaces (What is Spaces?)
[ ] Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
( ) Related Papers

Recommenders and Search Tools

[ ] Link to Influence Flower
Influence Flower (What are Influence Flowers?)
[ ] Connected Papers Toggle
Connected Papers (What is Connected Papers?)
[ ] Core recommender toggle
CORE Recommender (What is CORE?)

  * Author
  * Venue
  * Institution
  * Topic

( ) 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.

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