mlsanityData to 14 Oct 2026

The Sensitivity of Word Embeddings-based Author Detection Models to Semantic-preserving Adversarial Perturbations

arXiv 2102.11917 ↗First posted 23 Feb 2021cs.CL cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Authorship analysis is an important subject in the field of natural language processing. It allows the detection of the most likely writer of articles, news, books, or messages. This technique has multiple uses in tasks related to authorship attribution, detection of plagiarism, style analysis, sources of misinformation, etc. The focus of this paper is to explore the limitations and sensitiveness of established approaches to adversarial manipulations of inputs. To this end, and using those established techniques, we first developed an experimental frame-work for author detection and input perturbations. Next, we experimentally evaluated the performance of the authorship detection model to a collection of semantic-preserving adversarial perturbations of input narratives. Finally, we compare and analyze the effects of different perturbation strategies, input and model configurations, and the effects of these on the author detection model.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 47 references that are in the map. First posted on arXiv 23 Feb 2021.

Builds on

15 of 47 references.

References · 15

  1. 2017Attention Is All You Needcited by 60,361
  2. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,474
  3. 2016Enriching Word Vectors with Subword Informationcited by 1,583
  4. 2019EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Taskscited by 807
  5. 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 671
  6. 2018Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidatescited by 639
  7. 2016Diachronic Word Embeddings Reveal Statistical Laws of Semantic Changecited by 221
  8. 2018Contextual Augmentation: Data Augmentation by Words with Paradigmatic Relationscited by 201
  9. 2019The Limitations of Stylometry for Detecting Machine-Generated Fake Newscited by 63
  10. 2019Variational Pretraining for Semi-supervised Text Classificationcited by 41
  11. 2020Detecting and Characterizing Bots that Commit Codecited by 10
  12. 2020Author2Vec: A Framework for Generating User Embeddingcited by 5
3 more
  1. 2020Towards Improved Model Design for Authorship Identification: A Survey on Writing Style Understandingcited by 3
  2. 2019A Machine Learning Framework for Authorship Identification From Textscited by 2
  3. 2019An Innovative Word Encoding Method For Text Classification Using Convolutional Neural Networkcited by 1

Built on by

Not cited yet.