Visual Comparison of Language Model Adaptation

To appear in IEEE Transactions on Visualization and Computer Graphics. Figure 1: We present a workspace that enables the evaluation and comparison of adapters - lightweight alternatives for language model fine-tuning. After data pre-processing (e.g., embedding extraction), users can select pre-trained adapters, create explanations, and explore model differences through three types of visualizations: Concept Embedding Similarity, Concept Embedding Projection, and Concept Prediction Similarity. The explanations are provided for single models as well as model comparisons. For each explanation, we provide further explanation details, such as the word contexts as well as embedding vectors themselves. Abstract--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-consuming. Thus, adapters have recently been introduced as a lightweight alternative for model adaptation. They consist of a small set of task-specific parameters with a reduced training time and simple parameter composition. The simplicity of adapter training and composition comes along with new challenges, such as maintaining an overview of adapter properties and effectively comparing their produced embedding spaces. To help developers overcome these challenges, we provide a twofold contribution. First, in close collaboration with NLP researchers, we conducted a requirement analysis for an approach supporting adapter evaluation and detected, among others, the need for both intrinsic (i.e., embedding similaritybased) and extrinsic (i.e., prediction-based) explanation methods. Second, motivated by the gathered requirements, we designed a flexible visual analytics workspace that enables the comparison of adapter properties. In this paper, we discuss several design iterations and alternatives for interactive, comparative visual explanation methods. Our comparative visualizations show the differences in the adapted embedding vectors and prediction outcomes for diverse human-interpretable concepts (e.g., person names, human qualities).