mlsanityData to 14 Oct 2026

Adversarial Example Generation with Syntactically Controlled Paraphrase Networks

arXiv 1804.06059 ↗First posted 17 Apr 2018cs.CLcited by 286

Not on the map. Why?

Cited by 34 in LLM safety and security, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We propose syntactically controlled paraphrase networks (SCPNs) and use them to generate adversarial examples. Given a sentence and a target syntactic form (e.g., a constituency parse), SCPNs are trained to produce a paraphrase of the sentence with the desired syntax. We show it is possible to create training data for this task by first doing backtranslation at a very large scale, and then using a parser to label the syntactic transformations that naturally occur during this process. Such data allows us to train a neural encoder-decoder model with extra inputs to specify the target syntax. A combination of automated and human evaluations show that SCPNs generate paraphrases that follow their target specifications without decreasing paraphrase quality when compared to baseline (uncontrolled) paraphrase systems. Furthermore, they are more capable of generating syntactically adversarial examples that both (1) "fool" pretrained models and (2) improve the robustness of these models to syntactic variation when used to augment their training data.

Key facts. Cited by 286 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: 77; 2022: 50; 2023: 74; 2024: 49; 2025: 25; 2026: 11.

286 papers.

Sort:

By area:

  1. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,429
  2. 2023"Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Modelscited by 489
  3. 2021Recent Advances in Natural Language Processing via Large Pre-Trained Language Models: A Surveycited by 463
  4. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 429
  5. 2021A Survey of Data Augmentation Approaches for NLPcited by 409
  6. 2021Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Modelscited by 232
  7. 2021Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Modelscited by 222
  8. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Triggercited by 222
  9. 2023Survey of Vulnerabilities in Large Language Models Revealed by Adversarial Attackscited by 194
  10. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfercited by 158
  11. 2023Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMscited by 124
  12. 2023Instructions as Backdoors: Backdoor Vulnerabilities of Instruction Tuning for Large Language Modelscited by 110
38 more
  1. 2021Measure and Improve Robustness in NLP Models: A Surveycited by 107
  2. 2022Why Should Adversarial Perturbations be Imperceptible? Rethink the Research Paradigm in Adversarial NLPcited by 102
  3. 2021An Empirical Survey of Data Augmentation for Limited Data Learning in NLPcited by 101
  4. 2023A Survey of Safety and Trustworthiness of Large Language Models through the Lens of Verification and Validationcited by 93
  5. 2024Breaking Agents: Compromising Autonomous LLM Agents Through Malfunction Amplificationcited by 81
  6. 2023Evaluating Large Language Models on Controlled Generation Taskscited by 73
  7. 2023Unleashing the potential of prompt engineering for large language modelscited by 72
  8. 2021Bad Characters: Imperceptible NLP Attackscited by 71
  9. 2021Towards Robustness Against Natural Language Word Substitutionscited by 69
  10. 2021Tailor: Generating and Perturbing Text with Semantic Controlscited by 68
  11. 2023VLATTACK: Multimodal Adversarial Attacks on Vision-Language Tasks via Pre-trained Modelscited by 68
  12. 2023Black Box Adversarial Prompting for Foundation Modelscited by 62
  13. 2024Instruction Backdoor Attacks Against Customized LLMscited by 62
  14. 2023Red Teaming Language Model Detectors with Language Modelscited by 55
  15. 2021Explanation-Based Human Debugging of NLP Models: A Surveycited by 53
  16. 2022BITE: Textual Backdoor Attacks with Iterative Trigger Injectioncited by 53
  17. 2023Dr.Spider: A Diagnostic Evaluation Benchmark towards Text-to-SQL Robustnesscited by 53
  18. 2024Semantic Mirror Jailbreak: Genetic Algorithm Based Jailbreak Prompts Against Open-source LLMscited by 53
  19. 2021CLINE: Contrastive Learning with Semantic Negative Examples for Natural Language Understandingcited by 50
  20. 2024No Free Lunch in LLM Watermarking: Trade-offs in Watermarking Design Choicescited by 50
  21. 2021How Should Pre-Trained Language Models Be Fine-Tuned Towards Adversarial Robustness?cited by 49
  22. 2022VALUE: Understanding Dialect Disparity in NLUcited by 47
  23. 2021Generating Syntactically Controlled Paraphrases without Using Annotated Parallel Pairscited by 45
  24. 2023The language of prompting: What linguistic properties make a prompt successful?cited by 44
  25. 2023On Robustness of Prompt-based Semantic Parsing with Large Pre-trained Language Model: An Empirical Study on Codexcited by 41
  26. 2023In ChatGPT We Trust? Measuring and Characterizing the Reliability of ChatGPTcited by 37
  27. 2023Are Personalized Stochastic Parrots More Dangerous? Evaluating Persona Biases in Dialogue Systemscited by 37
  28. 2021Controllable Neural Dialogue Summarization with Personal Named Entity Planningcited by 36
  29. 2022Can Rationalization Improve Robustness?cited by 36
  30. 2024Best-of-Venom: Attacking RLHF by Injecting Poisoned Preference Datacited by 35
  31. 2023Text-CRS: A Generalized Certified Robustness Framework against Textual Adversarial Attackscited by 34
  32. 2023Attention-Enhancing Backdoor Attacks Against BERT-based Modelscited by 32
  33. 2022Quality Controlled Paraphrase Generationcited by 31
  34. 2021Defending Pre-trained Language Models from Adversarial Word Substitutions Without Performance Sacrificecited by 30
  35. 2021Perturbing Inputs for Fragile Interpretations in Deep Natural Language Processingcited by 30
  36. 2023Assessing Hidden Risks of LLMs: An Empirical Study on Robustness, Consistency, and Credibilitycited by 30
  37. 2021RockNER: A Simple Method to Create Adversarial Examples for Evaluating the Robustness of Named Entity Recognition Modelscited by 28
  38. 2021Bridge the Gap Between CV and NLP! A Gradient-based Textual Adversarial Attack Frameworkcited by 28

50 most cited of 286 All 286 (JSON)