Corpus ID: 203951318

The Daunting Task of Real-World Textual Style Transfer Auto-Evaluation

  title={The Daunting Task of Real-World Textual Style Transfer Auto-Evaluation},
  author={Richard Yuanzhe Pang},
  • Richard Yuanzhe Pang
  • Published 2019
  • Computer Science
  • ArXiv
  • The difficulty of textual style transfer lies in the lack of parallel corpora. Numerous advances have been proposed for the unsupervised generation. However, significant problems remain with the auto-evaluation of style transfer tasks. Based on the summary of Pang and Gimpel (2018) and Mir et al. (2019), style transfer evaluations rely on three criteria: style accuracy of transferred sentences, content similarity between original and transferred sentences, and fluency of transferred sentences… CONTINUE READING

    Tables and Topics from this paper.

    Explore Further: Topics Discussed in This Paper

    Review of Text Style Transfer Based on Deep Learning
    • 1
    • PDF


    Publications referenced by this paper.
    Style Transfer Through Back-Translation
    • 161
    • PDF
    Style Transfer in Text: Exploration and Evaluation
    • 209
    • PDF
    Style Transfer from Non-Parallel Text by Cross-Alignment
    • 337
    • Highly Influential
    • PDF
    Bleu: a Method for Automatic Evaluation of Machine Translation
    • 12,020
    • Highly Influential
    • PDF
    Glove: Global Vectors for Word Representation
    • 14,585
    • PDF
    Toward Controlled Generation of Text
    • 450
    • Highly Influential
    • PDF
    Evaluating Style Transfer for Text
    • 20
    • Highly Influential
    • PDF
    Unsupervised Text Style Transfer using Language Models as Discriminators
    • 116
    • PDF
    Data-Driven Response Generation in Social Media
    • 490
    • PDF