https://arxiv.org/abs/2307.04721

Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member
institutions, and all contributors. Donate
 
arxiv logo > cs > arXiv:2307.04721
[                    ]

Help | Advanced Search

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

  * Login
  * Help Pages
  * About

Computer Science > Artificial Intelligence

arXiv:2307.04721 (cs)
[Submitted on 10 Jul 2023]

Title:Large Language Models as General Pattern Machines

Authors:Suvir Mirchandani, Fei Xia, Pete Florence, Brian Ichter,
Danny Driess, Montserrat Gonzalez Arenas, Kanishka Rao, Dorsa Sadigh,
Andy Zeng
Download a PDF of the paper titled Large Language Models as General
Pattern Machines, by Suvir Mirchandani and 8 other authors
Download PDF

    Abstract: We observe that pre-trained large language models
    (LLMs) are capable of autoregressively completing complex token
    sequences -- from arbitrary ones procedurally generated by
    probabilistic context-free grammars (PCFG), to more rich spatial
    patterns found in the Abstract Reasoning Corpus (ARC), a general
    AI benchmark, prompted in the style of ASCII art. Surprisingly,
    pattern completion proficiency can be partially retained even
    when the sequences are expressed using tokens randomly sampled
    from the vocabulary. These results suggest that without any
    additional training, LLMs can serve as general sequence modelers,
    driven by in-context learning. In this work, we investigate how
    these zero-shot capabilities may be applied to problems in
    robotics -- from extrapolating sequences of numbers that
    represent states over time to complete simple motions, to
    least-to-most prompting of reward-conditioned trajectories that
    can discover and represent closed-loop policies (e.g., a
    stabilizing controller for CartPole). While difficult to deploy
    today for real systems due to latency, context size limitations,
    and compute costs, the approach of using LLMs to drive low-level
    control may provide an exciting glimpse into how the patterns
    among words could be transferred to actions.

Comments: 18 pages, 23 figures
Subjects: Artificial Intelligence (cs.AI); Computation and Language
          (cs.CL); Robotics (cs.RO)
Cite as:  arXiv:2307.04721 [cs.AI]
          (or arXiv:2307.04721v1 [cs.AI] for this version)
          https://doi.org/10.48550/arXiv.2307.04721
          Focus to learn more
          arXiv-issued DOI via DataCite

Submission history

From: Suvir Mirchandani [view email]
[v1] Mon, 10 Jul 2023 17:32:13 UTC (7,581 KB)
Full-text links:

Download:

  * Download a PDF of the paper titled Large Language Models as
    General Pattern Machines, by Suvir Mirchandani and 8 other
    authors
    PDF
  * Other formats

[by-4]
Current browse context:
cs.AI
< prev   |   next >
new | recent | 2307
Change to browse by:
cs
cs.CL
cs.RO

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?)
[ ] 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?)
( ) 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