What is wrong with style transfer for texts?
arXiv:1808.04365v127 citations
Originality Synthesis-oriented
AI Analysis
This incremental work aims to clarify a foundational problem in NLP for researchers in text style transfer.
The paper identifies a lack of consensus in automated text style transfer, with researchers using varied algorithms, datasets, and metrics, and proposes a formalization to address this issue.
A number of recent machine learning papers work with an automated style transfer for texts and, counter to intuition, demonstrate that there is no consensus formulation of this NLP task. Different researchers propose different algorithms, datasets and target metrics to address it. This short opinion paper aims to discuss possible formalization of this NLP task in anticipation of a further growing interest to it.