• Mapping the Latent Spaces of Culture

    Author(s):
    Ted Underwood (see profile)
    Date:
    2021
    Group(s):
    Digital Humanists, Sociology, TC Digital Humanities
    Subject(s):
    Machine learning, Natural language processing (Computer science), Culture--Philosophy
    Item Type:
    Article
    Tag(s):
    Natural language processing, Cultural theory
    Permanent URL:
    http://dx.doi.org/10.17613/faaa-1r21
    Abstract:
    As neural language models begin to change aspects of everyday life, they understandably attract criticism. This position paper was commissioned for a roundtable at Princeton University, dedicated to one of the most influential critiques: "On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?" by Emily M. Bender, Timnit Gebru, Angelina McMillan-Major, and Margaret Mitchell. My paper agrees that neural language models pose a variety of dangers, starting with and not limited to the list in "Stochastic Parrots." But to understand those dangers, I think we need to look beyond the premise that these models mimic "language understanding" on an individual level. That may have been what linguists and computer scientists intended them to do. But the models' actual potential (for both good and ill) is more interesting, and will be easier to grasp if we approach them as models of culture. Science-fictional scenarios about robots that become autonomous (or remain mere "parrots") are less useful here than humanistic cultural theory.
    Notes:
    This is a preprint version of a text that may subsequently appear in _Startwords_.
    Metadata:
    Status:
    Published
    Last Updated:
    1 year ago
    License:
    Attribution
    Share this:

    Downloads

    Item Name: pdf mappingthelatentspacesofculture.pdf
      Download View in browser
    Activity: Downloads: 241