mlsanityData to 14 Oct 2026

Delete, Retrieve, Generate: A Simple Approach to Sentiment and Style Transfer

arXiv 1804.06437 ↗First posted 17 Apr 2018cs.CLcited by 191

Not on the map. Why?

Abstract (arXiv)

We consider the task of text attribute transfer: transforming a sentence to alter a specific attribute (e.g., sentiment) while preserving its attribute-independent content (e.g., changing "screen is just the right size" to "screen is too small"). Our training data includes only sentences labeled with their attribute (e.g., positive or negative), but not pairs of sentences that differ only in their attributes, so we must learn to disentangle attributes from attribute-independent content in an unsupervised way. Previous work using adversarial methods has struggled to produce high-quality outputs. In this paper, we propose simpler methods motivated by the observation that text attributes are often marked by distinctive phrases (e.g., "too small"). Our strongest method extracts content words by deleting phrases associated with the sentence's original attribute value, retrieves new phrases associated with the target attribute, and uses a neural model to fluently combine these into a final output. On human evaluation, our best method generates grammatical and appropriate responses on 22% more inputs than the best previous system, averaged over three attribute transfer datasets: altering sentiment of reviews on Yelp, altering sentiment of reviews on Amazon, and altering image captions to be more romantic or humorous.

Key facts. Cited by 191 papers in the map (data through 14 Oct 2026). First posted on arXiv 17 Apr 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 59; 2022: 49; 2023: 37; 2024: 33; 2025: 9; 2026: 4.

191 papers.

Sort:

By area:

  1. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  2. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,195
  3. 2023Steering Language Models With Activation Engineeringcited by 922
  4. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 441
  5. 2022RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learningcited by 366
  6. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfercited by 158
  7. 2022A Survey on Retrieval-Augmented Text Generationcited by 158
  8. 2021A Recipe For Arbitrary Text Style Transfer with Large Language Modelscited by 107
  9. 2021The Perils of Using Mechanical Turk to Evaluate Open-Ended Text Generationcited by 97
  10. 2023Controlled Text Generation with Natural Language Instructionscited by 87
  11. 2023Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Modelcited by 79
  12. 2021Towards Facilitating Empathic Conversations in Online Mental Health Support: A Reinforcement Learning Approachcited by 75
38 more
  1. 2021Controlled Text Generation as Continuous Optimization with Multiple Constraintscited by 70
  2. 2023Word Embeddings Are Steers for Language Modelscited by 70
  3. 2022Latent Diffusion Energy-Based Model for Interpretable Text Modelingcited by 67
  4. 2023Detecting and Grounding Multi-Modal Media Manipulation and Beyondcited by 56
  5. 2021XFORMAL: A Benchmark for Multilingual Formality Style Transfercited by 52
  6. 2021Text Detoxification using Large Pre-trained Neural Modelscited by 52
  7. 2022Prompt-and-Rerank: A Method for Zero-Shot and Few-Shot Arbitrary Textual Style Transfer with Small Language Modelscited by 49
  8. 2023Teach LLMs to Personalize -- An Approach inspired by Writing Educationcited by 48
  9. 2022Less is More: Learning to Refine Dialogue History for Personalized Dialogue Generationcited by 47
  10. 2022A Causal Lens for Controllable Text Generationcited by 45
  11. 2023Benchmarking and Improving Generator-Validator Consistency of Language Modelscited by 44
  12. 2021LEWIS: Levenshtein Editing for Unsupervised Text Style Transfercited by 40
  13. 2021A Novel Estimator of Mutual Information for Learning to Disentangle Textual Representationscited by 37
  14. 2022Detoxifying Text with MaRCo: Controllable Revision with Experts and Anti-Expertscited by 36
  15. 2021StylePTB: A Compositional Benchmark for Fine-grained Controllable Text Style Transfercited by 34
  16. 2021Civil Rephrases Of Toxic Texts With Self-Supervised Transformerscited by 32
  17. 2022Meet Your Favorite Character: Open-domain Chatbot Mimicking Fictional Characters with only a Few Utterancescited by 32
  18. 2022Inducing Positive Perspectives with Text Reframingcited by 31
  19. 2022Text Style Transfer for Bias Mitigation using Masked Language Modelingcited by 28
  20. 2022Composable Text Controls in Latent Space with ODEscited by 28
  21. 2021Parallel Refinements for Lexically Constrained Text Generation with BARTcited by 24
  22. 2022APPDIA: A Discourse-aware Transformer-based Style Transfer Model for Offensive Social Media Conversationscited by 23
  23. 2024On the Evaluation of Machine-Generated Reportscited by 23
  24. 2021RetGen: A Joint framework for Retrieval and Grounded Text Generation Modelingcited by 22
  25. 2021Few-shot Controllable Style Transfer for Low-Resource Multilingual Settingscited by 21
  26. 2024A Survey on Natural Language Counterfactual Generationcited by 21
  27. 2021Enhancing Content Preservation in Text Style Transfer Using Reverse Attention and Conditional Layer Normalizationcited by 20
  28. 2022Controllable Data Generation by Deep Learning: A Reviewcited by 20
  29. 2021A Review of Human Evaluation for Style Transfercited by 19
  30. 2022Why is constrained neural language generation particularly challenging?cited by 19
  31. 2021A Review of Text Style Transfer using Deep Learningcited by 18
  32. 2022FRMT: A Benchmark for Few-Shot Region-Aware Machine Translationcited by 18
  33. 2023Evade ChatGPT Detectors via A Single Spacecited by 18
  34. 2021Don't Take It Literally: An Edit-Invariant Sequence Loss for Text Generationcited by 17
  35. 2022Text Revision by On-the-Fly Representation Optimizationcited by 17
  36. 2022Proactively Reducing the Hate Intensity of Online Posts via Hate Speech Normalizationcited by 17
  37. 2023Can Authorship Representation Learning Capture Stylistic Features?cited by 17
  38. 2023Prompt-Based Editing for Text Style Transfercited by 16

50 most cited of 191 All 191 (JSON)