mlsanityData to 7 Oct 2026

Defense Against Syntactic Textual Backdoor Attacks with Token Substitution

arXiv 2407.04179 ↗First posted 4 Jul 2024cs.CLcited by 1

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

Abstract (arXiv)

Textual backdoor attacks present a substantial security risk to Large Language Models (LLM). It embeds carefully chosen triggers into a victim model at the training stage, and makes the model erroneously predict inputs containing the same triggers as a certain class. Prior backdoor defense methods primarily target special token-based triggers, leaving syntax-based triggers insufficiently addressed. To fill this gap, this paper proposes a novel online defense algorithm that effectively counters syntax-based as well as special token-based backdoor attacks. The algorithm replaces semantically meaningful words in sentences with entirely different ones but preserves the syntactic templates or special tokens, and then compares the predicted labels before and after the substitution to determine whether a sentence contains triggers. Experimental results confirm the algorithm's performance against these two types of triggers, offering a comprehensive defense strategy for model integrity.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security. Builds on 41 of its 60 references that are in the map. First posted on arXiv 4 Jul 2024.

Builds on

41 of 60 references.

Closest ancestors · 15

Same direction, newest first

  1. 2023Backdoor Attacks for In-Context Learning with Language Models27 Jul
  2. 2023Defending against Insertion-based Textual Backdoor Attacks via Attribution3 May
  3. 2023ChatGPT as an Attack Tool: Stealthy Textual Backdoor Attack via Blackbox Generative Model Trigger27 Apr
  4. 2023Detecting Backdoors in Pre-trained Encoders23 Mar
  5. 2023Training-free Lexical Backdoor Attacks on Language Models8 Feb
  6. 2023BDMMT: Backdoor Sample Detection for Language Models through Model Mutation Testing25 Jan
  7. 2022BadPrompt: Backdoor Attacks on Continuous Prompts27 Nov
  8. 2022Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor Attacks14 Oct
  9. 2022A Unified Evaluation of Textual Backdoor Learning: Frameworks and Benchmarks17 Jun
  10. 2022BITE: Textual Backdoor Attacks with Iterative Trigger Injection25 May
  11. 2021RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Models15 Oct
  12. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer14 Oct
3 more
  1. 2021Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution11 Jun
  2. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger26 May
  3. 2021T-Miner: A Generative Approach to Defend Against Trojan Attacks on DNN-based Text Classification7 Mar

Other references · 26

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2023GPT-4 Technical Reportcited by 18,132
  4. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  5. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  6. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,243
  7. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,689
  8. 2015Character-level Convolutional Networks for Text Classificationcited by 2,293
  9. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
  10. 2020Extracting Training Data from Large Language Modelscited by 2,144
  11. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  12. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
14 more
  1. 2019Universal Adversarial Triggers for Attacking and Analyzing NLPcited by 709
  2. 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 568
  3. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 550
  4. 2020Invisible Backdoor Attack with Sample-Specific Triggerscited by 338
  5. 2020Weight Poisoning Attacks on Pre-trained Modelscited by 318
  6. 2020Input-Aware Dynamic Backdoor Attackcited by 304
  7. 2020ONION: A Simple and Effective Defense Against Textual Backdoor Attackscited by 231
  8. 2019A backdoor attack against LSTM-based text classification systemscited by 214
  9. 2024Security and Privacy Challenges of Large Language Models: A Surveycited by 179
  10. 2017Natural Language Processing: State of The Art, Current Trends and Challengescited by 114
  11. 2020Mitigating backdoor attacks in LSTM-based Text Classification Systems by Backdoor Keyword Identificationcited by 80
  12. 2019Design and Evaluation of a Multi-Domain Trojan Detection Method on Deep Neural Networkscited by 66
  13. 2022A Study of the Attention Abnormality in Trojaned BERTscited by 47
  14. 2024LOTUS: Evasive and Resilient Backdoor Attacks through Sub-Partitioningcited by 13

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025BadJudge: Backdoor Vulnerabilities of LLM-as-a-Judgecited by 13