LMU's Center for Advanced Studies (CAS) invites university members to a CAS Lunch Talk with Prof. Ted Underwood, Ph.D. (CAS Fellow/Digital Humanities, University of Illinois).
Abstract:
Researchers have worked hard over the last quarter-century to stretch interpretive disciplines toward larger scales of analysis. Ironically, we are now confronted with a converse problem. The language models increasingly ubiquitous in our lives are typically trained on the largest possible context: almost everything their creators can find. This breadth has value, but models unfortunately turn out to be quite bad at ignoring things. So seeing a social panorama doesn't necessarily give them a human-like ability to focus on one part, understanding that situation from the inside, while tuning out other places, times, and social positions. When models try that, they often produce a caricature that is still seen from the outside. In this lunch talk, I'll discuss various strategies for scaling context down, and also describe a benchmark that aims to tell us whether models have scaled down successfully.
The event is part of the CAS Research Focus “Scales” and chaired by Prof. Dr. Julian Schröter (LMU Digital Literary Studies). Following the talk and discussion, participants are invited to a light lunch at CAS.
Registration is required via email: info@cas.lmu.de