No article text was recorded for this version.
word-substitution vulnerability
Version 1 of 192 papers100 conceptsfrozen 2026-10-04
Papers in this version
- Blog: CheckHARD: Checking Hard Labels for Adversarial Text Detection, Prediction Correction, and Perturbed Word Suggestion
- Blog: Word Level Robustness Enhancement: Fight Perturbation with Perturbation
- Blog: Natural language adversarial defense through synonym encoding
- Blog: Defense of Word-Level Adversarial Attacks via Random Substitution Encoding
- Blog: Stability Analysis of ChatGPT-based Sentiment Analysis in AI Quality Assurance
- Blog: Towards Realistic Guarantees: A Probabilistic Certificate for SmoothLLM
- Blog: A Geometrical Approach to Finding Difficult Examples in Language
- Blog: Language Model Pre-training on True Negatives
- Blog: Value Decomposition Fails in Anti-Coordination: A Systematic MARL Comparison for Dynamic Spectrum Access in Dense 6G Networks
- Blog: SAFER: A Structure-free Approach for Certified Robustness to Adversarial Word Substitutions
- Blog: A View on Vulnerabilites: The Security Challenges of XAI (Academic Track)
- Blog: CT-GAT: Cross-Task Generative Adversarial Attack based on Transferability
- Blog: SemRoDe: Macro Adversarial Training to Learn Representations that are Robust to Word-Level Attacks
- Blog: Synonym-unaware Fast Adversarial Training against Textual Adversarial Attacks
- Blog: Don't Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text
- Blog: Text Adversarial Purification as Defense against Adversarial Attacks
- Blog: Misleading Relation Classifiers by Substituting Words in Texts
- Blog: Evaluating and Enhancing the Robustness of Neural Network-based Dependency Parsing Models with Adversarial Examples
- Blog: Enhancing LLM Robustness to Perturbed Instructions: An Empirical Study
- Blog: Reasoning Inconsistencies and How to Mitigate Them in Deep Learning
- Blog: Bridging Robustness and Generalization Against Word Substitution Attacks in NLP via the Growth Bound Matrix Approach
- Blog: An LLM can Fool Itself: A Prompt-Based Adversarial Attack
- Blog: UniT: A Unified Look at Certified Robust Training against Text Adversarial Perturbation
- Blog: Language Model Pre-training on True Negatives
- Blog: TextShield: Beyond Successfully Detecting Adversarial Sentences in Text Classification
- Blog: Robustness-Aware Word Embedding Improves Certified Robustness to Adversarial Word Substitutions
- Blog: Less is More: Understanding Word-level Textual Adversarial Attack via n-gram Frequency Descend
- Blog: End-to-end Adversarial Sample Generation for Data Augmentation
- Blog: TrustGLM: Evaluating the Robustness of GraphLLMs Against Prompt, Text, and Structure Attacks
- Blog: Adversarial Training with Fast Gradient Projection Method against Synonym Substitution Based Text Attacks
- Blog: "That Is a Suspicious Reaction!": Interpreting Logits Variation to Detect NLP Adversarial Attacks
- Blog: Detecting textual adversarial examples through randomized substitution and vote
- Blog: In and Out-of-Domain Text Adversarial Robustness via Label Smoothing
- Blog: MINIMAL: Mining Models for Universal Adversarial Triggers
- Blog: Universal Rules for Fooling Deep Neural Networks based Text Classification
- Blog: LEAP: Efficient and Automated Test Method for NLP Software
- Blog: BERT Rankers are Brittle
- Blog: Certified Robustness Against Natural Language Attacks by Causal Intervention
- Blog: Text Adversarial Purification as Defense against Adversarial Attacks
- Blog: Self-Supervised Contrastive Learning with Adversarial Perturbations for Robust Pretrained Language Models
- Blog: Catch Me If You Can: Detecting Phishing Emails Through Generative-Adversarial Training
- Blog: The Best Defense is Attack: Repairing Semantics in Textual Adversarial Examples
- Blog: Syntactically-Informed Unsupervised Paraphrasing with Non-Parallel Data
- Blog: FATRER: Full-Attention Topic Regularizer for Accurate and Robust Conversational Emotion Recognition
- Blog: Towards Self-Robust LLMs: Intrinsic Prompt Noise Resistance via CoIPO
- Blog: Mere Contrastive Learning for Cross-Domain Sentiment Analysis
- Blog: Natural Language Adversarial Attacks and Defenses in Word Level
- Blog: Natural Language Adversarial Defense through Synonym Encoding
- Blog: MPAT: Building Robust Deep Neural Networks against Textual Adversarial Attacks
- Blog: Self-Supervised Contrastive Learning with Adversarial Perturbations for Defending Word Substitution-based Attacks
- Blog: CAT-Gen: Improving Robustness in NLP Models via Controlled Adversarial Text Generation
- Blog: Adversarial Training for Improving Model Robustness? Look at Both Prediction and Interpretation
- Blog: pair2vec: Compositional Word-Pair Embeddings for Cross-Sentence Inference
- Blog: Did the Model Understand the Question?
- Blog: Robust textual embedding against word-level adversarial attacks
- Blog: Towards Adversarially Robust Text Classifiers by Learning to Reweight Clean Examples
- Blog: Crafting Adversarial Examples for Neural Machine Translation
- Blog: Frequency-Guided Word Substitutions for Detecting Textual Adversarial Examples
- Blog: Tougher Text, Smarter Models: Raising the Bar for Adversarial Defence Benchmarks
- Blog: How Does Context Matter? On the Robustness of Event Detection with Context-Selective Mask Generalization
Showing 60 of 92 papers.
History & citation
Current · Permanent link
- Robustness to Synonym Substitution in Natural Language Processing merged into this direction (75% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_word_substitution_vulnerability_dbd47f99,
title = {word-substitution vulnerability},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/word-substitution-vulnerability/rdr_dbd47f99c40a44be9015d71c73d4c58b},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 1, frozen text):
@misc{lacuna_word_substitution_vulnerability_dbd47f99_v1,
title = {word-substitution vulnerability},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/word-substitution-vulnerability/rdr_dbd47f99c40a44be9015d71c73d4c58b/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}