Team ÚFAL at CMCL 2022 Shared Task: Figuring out the correct recipe for predicting Eye-Tracking features using Pretrained Language Models
This work addresses improving eye-tracking prediction for cognitive and language comprehension studies, but it is incremental as it builds on existing models and methods.
The paper tackled predicting eye-tracking features using pretrained language models like BERT and XLM, achieving an average MAE of 5.72 in a shared task and reducing it to 5.25 in post-evaluation.
Eye-Tracking data is a very useful source of information to study cognition and especially language comprehension in humans. In this paper, we describe our systems for the CMCL 2022 shared task on predicting eye-tracking information. We describe our experiments with pretrained models like BERT and XLM and the different ways in which we used those representations to predict four eye-tracking features. Along with analysing the effect of using two different kinds of pretrained multilingual language models and different ways of pooling the tokenlevel representations, we also explore how contextual information affects the performance of the systems. Finally, we also explore if factors like augmenting linguistic information affect the predictions. Our submissions achieved an average MAE of 5.72 and ranked 5th in the shared task. The average MAE showed further reduction to 5.25 in post task evaluation.