From Theories on Styles to their Transfer in Text: Bridging the Gap with a Hierarchical Survey

Abstract

Humans are naturally endowed with the ability to write in a particular style. They can, for instance, re-phrase a formal letter in an informal way, convey a literal message with the use of figures of speech or edit a novel by mimicking the style of some well-known authors. Automating this form of creativity constitutes the goal of style transfer. As a natural language generation task, style transfer aims at rewriting existing texts, and specifically, it creates paraphrases that exhibit some desired stylistic attributes. From a practical perspective, it envisions beneficial applications, like chatbots that modulate their communicative style to appear empathetic, or systems that automatically simplify technical articles for a non-expert audience. Several style-aware paraphrasing methods have attempted to tackle style transfer. A handful of surveys give a methodological overview of the field, but they do not support researchers to focus on specific styles. With this paper, we aim at providing a comprehensive discussion of the styles that have received attention in the transfer task. We organize them in a hierarchy, highlighting the challenges for the definition of each of them, and pointing out gaps in the current research landscape. The hierarchy comprises two main groups. One encompasses styles that people modulate arbitrarily, along the lines of registers and genres. The other group corresponds to unintentionally expressed styles, due to an author's personal characteristics. Hence, our review shows how these groups relate to one another, and where specific styles, including some that have not yet been explored, belong in the hierarchy. Moreover, we summarize the methods employed for different stylistic families, hinting researchers towards those that would be the most fitting for future research.

arXiv commentThis is a preprint of the article available open access at the Journal of Natural Language Processing. The final paper can be accessed via https://www.doi.org/10.1017/S1351324922000407 It has 58 pages, 2 figures, 10 tables. Please only cite the published version from NLE

References in corpus (33)

• BERTScore: Evaluating Text Generation with BERT
• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• ConceptNet 5.5: An Open Multilingual Graph of General Knowledge
• DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset
• Style Transfer from Non-Parallel Text by Cross-Alignment
• Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm
• Gated Feedback Recurrent Neural Networks
• A Computational Approach to Politeness with Application to Social Factors
• Content preserving text generation with attribute controls
• Multi-Task Neural Models for Translating Between Styles Within and Across Languages
• A Probabilistic Formulation of Unsupervised Text Style Transfer
• Unpaired Sentiment-to-Sentiment Translation: A Cycled Reinforcement Learning Approach
• FUDGE: Controlled Text Generation With Future Discriminators
• Unsupervised Text Style Transfer using Language Models as Discriminators
• Hate Towards the Political Opponent: A Twitter Corpus Study of the 2020 US Elections on the Basis of Offensive Speech and Stance Detection
• A Review of Text Style Transfer using Deep Learning
• A Dual Reinforcement Learning Framework for Unsupervised Text Style Transfer
• Evaluating prose style transfer with the Bible
• Style Transfer for Texts: Retrain, Report Errors, Compare with Rewrites
• Formality Style Transfer with Hybrid Textual Annotations
• Structured Content Preservation for Unsupervised Text Style Transfer
• Multi-Task Learning with Sentiment, Emotion, and Target Detection to Recognize Hate Speech and Offensive Language
• Decomposing Textual Information For Style Transfer
• Keep Calm and Switch On! Preserving Sentiment and Fluency in Semantic Text Exchange
• Zero-Shot Fine-Grained Style Transfer: Leveraging Distributed Continuous Style Representations to Transfer To Unseen Styles
• What is wrong with style transfer for texts?
• Sentiment Transfer using Seq2Seq Adversarial Autoencoders
• Domain Adaptive Text Style Transfer
• Stable Style Transformer: Delete and Generate Approach with Encoder-Decoder for Text Style Transfer
• The Daunting Task of Real-World Textual Style Transfer Auto-Evaluation
• A Neural Approach to Irony Generation
• Emotion Recognition under Consideration of the Emotion Component Process Model
• Style Transfer Through Multilingual and Feedback-Based Back-Translation