Abstract: The NLP task of text style transfer (TST) aims to automatically control the style attributes of a piece of text while preserving the content, which is an important consideration for making NLP more user-centric. In this session, we will explore text style transfer through an applied use case — neutralizing subjectivity bias in free text. Along the way, we’ll describe our sequence-to-sequence modeling approach leveraging HuggingFace Transformers, and present a set of custom, reference-free evaluation metrics for quantifying model performance. Finally, we’ll conclude with a discussion of ethics centered around our Applied Machine Learning Prototype: Exploring Intelligent Writing Assistance.
Bio: Melanie is the Research Engineering Manager of Cloudera Fast Forward Labs, an applied machine learning research team within Cloudera. As a researcher and data scientist, she is passionate about democratizing machine learning by turning academic breakthroughs into useful and accessible applications, especially in the NLP space. With experience as a data scientist in multiple industries from hardware manufacturing to cybersecurity, she is a jack of all trades who loves to share what she’s learned. She is also an avid knitter and a reformed astrophysicist, holding a Ph.D. in astrophysics from the University of Minnesota.