Transformer models are revolutionizing machine learning, but their inner workings remain
mysterious. In this work, we present a new visualization technique designed to help …

Traditional deep learning interpretability methods which are suitable for model users cannot
explain network behaviors at the global level and are inflexible at providing fine-grained …

The Transformer architecture has two main non-embedding components: Attention and the
Feed Forward Network (FFN). Attention captures interdependencies between words …

Neural language models are widely used; however, their model parameters often need to be
adapted to the specific domains and tasks of an application, which is time-and resource …

Modern machine learning techniques commonly rely on complex, high-dimensional
embedding representations to capture underlying structure in the data and improve …

Over the past years, an increasing number of publications in information visualization,
especially within the field of visual analytics, have mentioned the term “embedding” when …

Complexity and network science are nowadays used, or at least invoked, in a variety of
scientific researchareas ranging from the analysis of financial systems to social structure and …

Data features and class probabilities are two main perspectives when, eg, evaluating model
results and identifying problematic items. Class probabilities represent the likelihood that …

Despite the success of contextualized language models on various NLP tasks, it is still
unclear what these models really learn. In this paper, we contribute to the current efforts of …

Interpretability methods aim to help users build trust in and understand the capabilities of
machine learning models. However, existing approaches often rely on abstract, complex …