Data to 14 Oct 2026

Interpretable Adversarial Perturbation in Input Embedding Space for Text

arXiv 1805.02917 ↗First posted 8 May 2018cs.LG cs.CL stat.MLcited by 67

Not on the map. Why?

Abstract (arXiv)

Following great success in the image processing field, the idea of adversarial training has been applied to tasks in the natural language processing (NLP) field. One promising approach directly applies adversarial training developed in the image processing field to the input word embedding space instead of the discrete input space of texts. However, this approach abandons such interpretability as generating adversarial texts to significantly improve the performance of NLP tasks. This paper restores interpretability to such methods by restricting the directions of perturbations toward the existing words in the input embedding space. As a result, we can straightforwardly reconstruct each input with perturbations to an actual text by considering the perturbations to be the replacement of words in the sentence while maintaining or even improving the task performance.

Key facts. Cited by 67 papers in the map (data through 14 Oct 2026). First posted on arXiv 8 May 2018.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2018: 1; 2019: 0; 2020: 0; 2021: 24; 2022: 17; 2023: 12; 2024: 9; 2025: 4.

67 papers.

Sort:

By area:

2021Towards Robustness Against Natural Language Word Substitutionscited by 69
2021Certified Robustness to Text Adversarial Attacks by Randomized [MASK]cited by 57
2021Adversarial Attacks On Multi-Agent Communicationcited by 49
2021Defending Against Backdoor Attacks in Natural Language Generationcited by 33
2021Bridge the Gap Between CV and NLP! A Gradient-based Textual Adversarial Attack Frameworkcited by 28
2022Character-level White-Box Adversarial Attacks against Transformers via Attachable Subwords Substitutioncited by 28
2021The Role of Explainability in Assuring Safety of Machine Learning in Healthcarecited by 26
2022SEAT: Stable and Explainable Attentioncited by 25
2021ML Attack Models: Adversarial Attacks and Data Poisoning Attackscited by 24
2022A Survey of Adversarial Defences and Robustness in NLPcited by 23
2024SATO: Stable Text-to-Motion Frameworkcited by 20
2021Interpreting Deep Learning Models in Natural Language Processing: A Reviewcited by 19

38 more

2023Hyperpolyglot LLMs: Cross-Lingual Interpretability in Token Embeddingscited by 19
2021Efficient Combinatorial Optimization for Word-level Adversarial Textual Attackcited by 17
2022Improving the Adversarial Robustness of NLP Models by Information Bottleneckcited by 17
2024LLMScan: Causal Scan for LLM Misbehavior Detectioncited by 13
2022Defending Substitution-Based Profile Pollution Attacks on Sequential Recommenderscited by 11
2023Toward Adversarial Training on Contextualized Language Representationcited by 10
2021A Differentiable Language Model Adversarial Attack on Text Classifierscited by 9
2022Block-Sparse Adversarial Attack to Fool Transformer-Based Text Classifierscited by 9
2022Interpretation of Black Box NLP Models: A Surveycited by 9
2021Evaluating Document Coherence Modellingcited by 8
2021Making Attention Mechanisms More Robust and Interpretable with Virtual Adversarial Trainingcited by 8
2022Adversarial Self-Attention for Language Understandingcited by 8
2021Adversarial Regularization as Stackelberg Game: An Unrolled Optimization Approachcited by 7
2021Improving the robustness and accuracy of biomedical language models through adversarial trainingcited by 7
2022TESTSGD: Interpretable Testing of Neural Networks Against Subtle Group Discriminationcited by 7
2021On The State of Data In Computer Vision: Human Annotations Remain Indispensable for Developing Deep Learning Modelscited by 6
2023Modeling Adversarial Attack on Pre-trained Language Models as Sequential Decision Makingcited by 6
2024Round Trip Translation Defence against Large Language Model Jailbreaking Attackscited by 6
2021Don't Search for a Search Method -- Simple Heuristics Suffice for Adversarial Text Attackscited by 5
2021Effective and Imperceptible Adversarial Textual Attack via Multi-objectivizationcited by 5
2022Software Testing for Machine Learningcited by 5
2022Towards Responsible AI for Financial Transactionscited by 5
2024ProTransformer: Robustify Transformers via Plug-and-Play Paradigmcited by 5
2021Adversarial Attacks and Defenses for Social Network Text Processing Applications: Techniques, Challenges and Future Research Directionscited by 4
2022Improving robustness of language models from a geometry-aware perspectivecited by 4
2023Deep Learning-based Sentiment Classification: A Comparative Surveycited by 4
2021TREATED:Towards Universal Defense against Textual Adversarial Attackscited by 3
2022FuncFooler: A Practical Black-box Attack Against Learning-based Binary Code Similarity Detection Methodscited by 3
2023Adversary for Social Good: Leveraging Adversarial Attacks to Protect Personal Attribute Privacycited by 3
2023NatLogAttack: A Framework for Attacking Natural Language Inference Models with Natural Logiccited by 3
2024SemRoDe: Macro Adversarial Training to Learn Representations That are Robust to Word-Level Attackscited by 3
2024Enhancing Adversarial Text Attacks on BERT Models with Projected Gradient Descentcited by 3
2018Mixture of Expert/Imitator Networks: Scalable Semi-supervised Learning Frameworkcited by 2
2021Adversarial Machine Learning in Text Analysis and Generationcited by 2
2021Improving Robustness of Malware Classifiers using Adversarial Strings Generated from Perturbed Latent Representationscited by 2
2022Quantifying Robustness to Adversarial Word Substitutionscited by 2
2022Semantic-Preserving Adversarial Code Comprehensioncited by 2
2024TF-Attack: Transferable and Fast Adversarial Attacks on Large Language Modelscited by 2

50 most cited of 67 All 67 (JSON)