A new multi-LLM framework that lets three AI models iteratively refine scoring rubrics before grading reached an average Quadratic Weighted Kappa of 0.71 on the ASAP essay dataset, peaking at 0.86.
A two-phase calibration study found AI matched expert raters well on concrete checklist items in simulated patient encounters ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results