mlsanityData to 14 Oct 2026

Efficient Trigger Word Insertion

arXiv 2311.13957 ↗First posted 23 Nov 2023cs.CR cs.CLcited by 2

On the map: LLM safety and security → Backdoor and data poisoning attacks

Abstract (arXiv)

With the boom in the natural language processing (NLP) field these years, backdoor attacks pose immense threats against deep neural network models. However, previous works hardly consider the effect of the poisoning rate. In this paper, our main objective is to reduce the number of poisoned samples while still achieving a satisfactory Attack Success Rate (ASR) in text backdoor attacks. To accomplish this, we propose an efficient trigger word insertion strategy in terms of trigger word optimization and poisoned sample selection. Extensive experiments on different datasets and models demonstrate that our proposed method can significantly improve attack effectiveness in text classification tasks. Remarkably, our approach achieves an ASR of over 90% with only 10 poisoned samples in the dirty-label setting and requires merely 1.5% of the training data in the clean-label setting.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security. Builds on 30 of its 35 references that are in the map. First posted on arXiv 23 Nov 2023.

Builds on

30 of 35 references.

Closest ancestors · 6

Same direction, newest first

  1. 2022Kallima: A Clean-label Framework for Textual Backdoor Attacks3 Jun
  2. 2022BITE: Textual Backdoor Attacks with Iterative Trigger Injection25 May
  3. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer14 Oct
  4. 2021Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution11 Jun
  5. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger26 May
  6. 2021Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Models29 Mar

Other references · 24

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  4. 2013Intriguing properties of neural networkscited by 5,155
  5. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  6. 2019A Comprehensive Survey on Transfer Learningcited by 1,160
  7. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
  8. 2019Universal Adversarial Triggers for Attacking and Analyzing NLPcited by 722
  9. 2020Pre-trained Models for Natural Language Processing: A Surveycited by 505
  10. 2019How to Fine-Tune BERT for Text Classification?cited by 411
  11. 2020Weight Poisoning Attacks on Pre-trained Modelscited by 341
  12. 2018Not All Samples Are Created Equal: Deep Learning with Importance Samplingcited by 328
12 more
  1. 2019A backdoor attack against LSTM-based text classification systemscited by 227
  2. 2020BadNL: Backdoor Attacks against NLP Models with Semantic-preserving Improvementscited by 211
  3. 2018Backdoor Embedding in Convolutional Neural Network Models via Invisible Perturbationcited by 134
  4. 2019Fine-tune BERT for Extractive Summarizationcited by 122
  5. 2020Trojaning Language Models for Fun and Profitcited by 90
  6. 2020A Systematic Survey of Regularization and Normalization in GANscited by 27
  7. 2021Enhancing Backdoor Attacks with Multi-Level MMD Regularizationcited by 15
  8. 2022Data-Efficient Backdoor Attackscited by 13
  9. 2020A New Perspective on Stabilizing GANs training: Direct Adversarial Trainingcited by 12
  10. 2023Efficient Backdoor Attacks for Deep Neural Networks in Real-world Scenarioscited by 7
  11. 2023Explore the Effect of Data Selection on Poison Efficiency in Backdoor Attackscited by 7
  12. 2023A Proxy Attack-Free Strategy for Practically Improving the Poisoning Efficiency in Backdoor Attackscited by 5

Built on by

Cited by year

Citing papers per year: 2024: 2.

2 papers.

  1. 2024Persistent Backdoor Attacks in Continual Learningcited by 13
  2. 2024Large Language Models are Good Attackers: Efficient and Stealthy Textual Backdoor Attackscited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2025Strategic Sample Selection for Improved Clean-Label Backdoor Attacks in Text Classification
  2. 2022Rethink the Evaluation for Attack Strength of Backdoor Attacks in Natural Language Processing
  3. 2023Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models
  4. 2021Triggerless Backdoor Attack for NLP Tasks with Clean Labels
  5. 2023Large Language Models Are Better Adversaries: Exploring Generative Clean-Label Backdoor Attacks Against Text Classifiers
  6. 2024A Study of Backdoors in Instruction Fine-tuned Language Models
  7. 2024SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
  8. 2023Defending against Insertion-based Textual Backdoor Attacks via Attribution