These experiments are designed to determine the effectiveness of various AES techniques in Chemistry along with attempting to give feedback on short-medium length questions.
This experiment focuses on the difference in the marking between 3 different pre-trained transformer based models in the task of AES.
This experiment is to create a much smaller encoder + classification head model and pre-train on AES and SAS datasets to determine its effectiveness at being trained on much smaller models.