Grading Machines

• In Event: Being Human or More than Human: Teaching and Learning with Humanity

Brief Overview

we analyze how the GPT LLM performs in evaluating a series of graduate-level (law school) exams. Our analysis suggests that such tools could be used for valuable tasks in the near future, such as reviewing grading for errors and bias,

Authors

• Kevin Lynn Cope, University of Virginia

• Jens Frankenreiter, ETH Zurich

• Scott Hirst, Boston University

• Dane Thorley, J. Reuben Clark Law School, Brigham Young University