T3-Vis: a visual analytic framework for Training and fine-Tuning Transformers in NLP

TL;DR

T3-Vis is a visual analytic framework designed to assist NLP researchers in training and fine-tuning Transformer models by providing interactive visualizations and analytical tools to understand model properties and behaviors.

Contribution

It introduces a novel visual analytic framework that offers insights into Transformer models' intrinsic properties during training and fine-tuning, enhancing interpretability and research efficiency.

Findings

Framework provides intuitive exploration of hidden states and attention.

User feedback indicates the framework's usefulness and potential for improvements.

Case studies demonstrate practical benefits in model analysis.

Abstract

Transformers are the dominant architecture in NLP, but their training and fine-tuning is still very challenging. In this paper, we present the design and implementation of a visual analytic framework for assisting researchers in such process, by providing them with valuable insights about the model's intrinsic properties and behaviours. Our framework offers an intuitive overview that allows the user to explore different facets of the model (e.g., hidden states, attention) through interactive visualization, and allows a suite of built-in algorithms that compute the importance of model components and different parts of the input sequence. Case studies and feedback from a user focus group indicate that the framework is useful, and suggest several improvements.