Adversarial Text

Adversarial text research focuses on creating and defending against text inputs designed to deceive natural language processing (NLP) models, often by subtly altering wording while maintaining semantic similarity to a human reader. Current research emphasizes developing more effective attack methods, particularly those leveraging multi-agent systems, reinforcement learning, and diffusion models, as well as improving defenses through techniques like adversarial training and noise augmentation. This field is crucial for enhancing the robustness and trustworthiness of NLP systems across diverse applications, from automated essay scoring to autonomous vehicle navigation and large language model safety.

Papers

February 5, 2024

Adversarial Text Purification: A Large Language Model Approach for Defense
Raha Moraffah, Shubh Khandelwal, Amrita Bhattacharjee, Huan Liu
Large Language Model Adversarial Text Adversarial Purification Adversarial Text Purification

February 1, 2024

What Does the Bot Say? Opportunities and Risks of Large Language Models in Social Media Bot Detection
Shangbin Feng, Herun Wan, Ningnan Wang, Zhaoxuan Tan, Minnan Luo, Yulia Tsvetkov
Social Medium Emerging Opportunity Risk Sensitive Adversarial Text Bot Generated Text Social Bot Detection Bot Detection Bot Detector

January 22, 2024

Adversarial speech for voice privacy protection from Personalized Speech generation
Shihao Chen, Liping Chen, Jie Zhang, KongAik Lee, Zhenhua Ling, Lirong Dai
Adversarial Perturbation Speech Synthesis Adversarial Text Speech Generation Voice Privacy Personalized Speech

January 18, 2024

Power in Numbers: Robust reading comprehension by finetuning with four adversarial sentences per example
Ariel Marcus
Adversarial Attack Real Power Numerical Data Adversarial Text Reading Comprehension Single Example Comprehension Model

December 12, 2023

Adversarial Semi-Supervised Domain Adaptation for Semantic Segmentation: A New Role for Labeled Target Samples
Marwa Kechaou, Mokhtar Z. Alaya, Romain Hérault, Gilles Gasso
Semantic Segmentation Domain Adaptation Semi Supervised Adversarial Text Adversarial Loss Multi Role Adversarial Domain Adaptation Pseudo Labeled Sample

December 8, 2023

SA-Attack: Improving Adversarial Transferability of Vision-Language Pre-training Models via Self-Augmentation
Bangyan He, Xiaojun Jia, Siyuan Liang, Tianrui Lou, Yang Liu, Xiaochun Cao
Adversarial Example Full Model Adversarial Text Adversarial Image Self Augmentation White Box Adversarial Attack

December 3, 2023

TranSegPGD: Improving Transferability of Adversarial Examples on Semantic Segmentation
Xiaojun Jia, Jindong Gu, Yihao Huang, Simeng Qin, Qing Guo, Yang Liu, Xiaochun Cao
Adversarial Attack Semantic Segmentation Adversarial Example Task Transferability Adversarial Text

November 14, 2023

SAIE Framework: Support Alone Isn't Enough -- Advancing LLM Training with Adversarial Remarks
Mengsay Loem, Masahiro Kaneko, Naoaki Okazaki
Commonsense Reasoning Customer Service Adversarial Text LLM Training Learner Model

October 31, 2023

Robust Safety Classifier for Large Language Models: Adversarial Prompt Shield
Jinhwa Kim, Ali Derakhshan, Ian G. Harris
Adversarial Attack Adversarial Example Adversarial Text Adversarial Prompt Safety Classifier

October 29, 2023

October 25, 2023

Dual Defense: Adversarial, Traceable, and Invisible Robust Watermarking against Face Swapping
Yunming Zhang, Dengpan Ye, Caiyun Xie, Long Tang, Chuanxi Chen, Ziyi Liu, Jiacheng Deng
Adversarial Text Forgery Detection Face Swapping Robust Watermark Invisible Watermark

October 12, 2023

Effects of Human Adversarial and Affable Samples on BERT Generalization
Aparna Elangovan, Jiayuan He, Yuan Li, Karin Verspoor
Training Data Adversarial Example Mixed Effect BERT Model Generalization Performance Stronger Generalizability Adversarial Text Adversarial Information Novel Sample

September 30, 2023

Human-Producible Adversarial Examples
David Khachaturov, Yue Gao, Ilia Shumailov, Robert Mullins, Ross Anderson, Kassem Fawaz
Adversarial Example Adversarial Text

July 27, 2023

Adversarial Sleeping Bandit Problems with Multiple Plays: Algorithm and Ranking Application
Jianjun Yuan, Wei Lee Woon, Ludovik Coba
Practical Algorithm Adversarial Text Bandit Algorithm Adversarial Loss Play Style Bandit Problem First Stage Ranker Arm Selection Online Recommender System

June 14, 2023

A Relaxed Optimization Approach for Adversarial Attacks against Neural Machine Translation Models
Sahar Sadrizadeh, Clément Barbier, Ljiljana Dolamic, Pascal Frossard
Adversarial Attack Adversarial Example Adversarial Text Neural Machine Translation Model

June 9, 2023

Towards a Robust Detection of Language Model Generated Text: Is ChatGPT that Easy to Detect?
Wissam Antoun, Virginie Mouilleron, Benoît Sagot, Djamé Seddah
Natural Language Processing ChatGPT Generated Conversation Adversarial Text Model Generated Robust Detection ChatGPT Generated ChatGPT Detector

June 7, 2023

Arabic Dysarthric Speech Recognition Using Adversarial and Signal-Based Augmentation
Massa Baali, Ibrahim Almakky, Shady Shehata, Fakhri Karray
Speech Recognition Adversarial Text Dysarthric Speech Arabic Speech

June 6, 2023

Revisiting the Trade-off between Accuracy and Robustness via Weight Distribution of Filters
Xingxing Wei, Shiji Zhao, Bo li
Adversarial Attack Native Robustness Adversarial Example Adversarial Robustness Adversarial Text Max Filter Weight Distribution Adversarial Traffic

June 2, 2023

VoteTRANS: Detecting Adversarial Text without Training by Voting on Hard Labels of Transformations
Hoang-Quoc Nguyen-Son, Seira Hidano, Kazuhide Fukushima, Shinsaku Kiyomoto, Isao Echizen
Adversarial Attack Deep Learning Model Training Data Adversarial Text Voter Participation Left Corner Transformation Adversarial Data Hard Label

Adversarial Text

Papers

Adversarial Text Purification: A Large Language Model Approach for Defense

What Does the Bot Say? Opportunities and Risks of Large Language Models in Social Media Bot Detection

Adversarial speech for voice privacy protection from Personalized Speech generation

Power in Numbers: Robust reading comprehension by finetuning with four adversarial sentences per example

Adversarial Semi-Supervised Domain Adaptation for Semantic Segmentation: A New Role for Labeled Target Samples

SA-Attack: Improving Adversarial Transferability of Vision-Language Pre-training Models via Self-Augmentation

TranSegPGD: Improving Transferability of Adversarial Examples on Semantic Segmentation

SAIE Framework: Support Alone Isn't Enough -- Advancing LLM Training with Adversarial Remarks

Robust Safety Classifier for Large Language Models: Adversarial Prompt Shield

Poisoning Retrieval Corpora by Injecting Adversarial Passages

BERT Lost Patience Won't Be Robust to Adversarial Slowdown

Dual Defense: Adversarial, Traceable, and Invisible Robust Watermarking against Face Swapping

Effects of Human Adversarial and Affable Samples on BERT Generalization

Human-Producible Adversarial Examples

Adversarial Sleeping Bandit Problems with Multiple Plays: Algorithm and Ranking Application

A Relaxed Optimization Approach for Adversarial Attacks against Neural Machine Translation Models

Towards a Robust Detection of Language Model Generated Text: Is ChatGPT that Easy to Detect?

Arabic Dysarthric Speech Recognition Using Adversarial and Signal-Based Augmentation

Revisiting the Trade-off between Accuracy and Robustness via Weight Distribution of Filters

VoteTRANS: Detecting Adversarial Text without Training by Voting on Hard Labels of Transformations