Visualizing attention zones in machine reading comprehension models
This work provides a tool for researchers to interpret attention in MRC models, but it is incremental as it applies existing visualization techniques to a specific domain.
The authors tackled the problem of understanding attention mechanisms in machine reading comprehension models by developing a pipeline to visualize attention zones across layers, which aids in model explainability.
The attention mechanism plays an important role in the machine reading comprehension (MRC) model. Here, we describe a pipeline for building an MRC model with a pretrained language model and visualizing the effect of each attention zone in different layers, which can indicate the explainability of the model. With the presented protocol and accompanying code, researchers can easily visualize the relevance of each attention zone in the MRC model. This approach can be generalized to other pretrained language models.