mlsanityData to 7 Oct 2026

Mitigating backdoor attacks in LSTM-based Text Classification Systems by Backdoor Keyword Identification

arXiv 2007.12070 ↗First posted 11 Jul 2020cs.CR cs.LG stat.MLcited by 80

Not on the map. Why?

Cited by 62 in LLM safety and security, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

It has been proved that deep neural networks are facing a new threat called backdoor attacks, where the adversary can inject backdoors into the neural network model through poisoning the training dataset. When the input containing some special pattern called the backdoor trigger, the model with backdoor will carry out malicious task such as misclassification specified by adversaries. In text classification systems, backdoors inserted in the models can cause spam or malicious speech to escape detection. Previous work mainly focused on the defense of backdoor attacks in computer vision, little attention has been paid to defense method for RNN backdoor attacks regarding text classification. In this paper, through analyzing the changes in inner LSTM neurons, we proposed a defense method called Backdoor Keyword Identification (BKI) to mitigate backdoor attacks which the adversary performs against LSTM-based text classification by data poisoning. This method can identify and exclude poisoning samples crafted to insert backdoor into the model from training data without a verified and trusted dataset. We evaluate our method on four different text classification datset: IMDB, DBpedia ontology, 20 newsgroups and Reuters-21578 dataset. It all achieves good performance regardless of the trigger sentences.

Key facts. Cited by 80 papers in the map (data through 7 Oct 2026). First posted on arXiv 11 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 16; 2023: 16; 2024: 26; 2025: 11; 2026: 4.

80 papers.

Sort:

By area:

  1. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Triggercited by 203
  2. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfercited by 147
  3. 2021RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Modelscited by 97
  4. 2022Wild Patterns Reloaded: A Survey of Machine Learning Security against Training Data Poisoningcited by 93
  5. 2023Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Modelscited by 87
  6. 2021Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitutioncited by 85
  7. 2023Use of LLMs for Illicit Purposes: Threats, Prevention Measures, and Vulnerabilitiescited by 76
  8. 2022A Unified Evaluation of Textual Backdoor Learning: Frameworks and Benchmarkscited by 70
  9. 2022Rickrolling the Artist: Injecting Backdoors into Text Encoders for Text-to-Image Synthesiscited by 59
  10. 2022Fine-mixing: Mitigating Backdoors in Fine-tuned Language Modelscited by 50
  11. 2024BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Modelscited by 50
  12. 2023Privacy in Large Language Models: Attacks, Defenses and Future Directionscited by 48
38 more
  1. 2022BITE: Textual Backdoor Attacks with Iterative Trigger Injectioncited by 47
  2. 2022Constrained Optimization with Dynamic Bound-scaling for Effective NLPBackdoor Defensecited by 44
  3. 2023Defending Pre-trained Language Models as Few-shot Learners against Backdoor Attackscited by 32
  4. 2021Defending Against Backdoor Attacks in Natural Language Generationcited by 30
  5. 2023From Shortcuts to Triggers: Backdoor Defense with Denoised PoEcited by 26
  6. 2024Large Model Based Agents: State-of-the-Art, Cooperation Paradigms, Security and Privacy, and Future Trendscited by 26
  7. 2022BAFFLE: Hiding Backdoors in Offline Reinforcement Learning Datasetscited by 24
  8. 2023Test-time Backdoor Mitigation for Black-Box Large Language Models with Defensive Demonstrationscited by 20
  9. 2024BadActs: A Universal Backdoor Defense in the Activation Spacecited by 19
  10. 2022Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor Attackscited by 18
  11. 2021BFClass: A Backdoor-free Text Classification Frameworkcited by 17
  12. 2023Data and Model Poisoning Backdoor Attacks on Wireless Federated Learning, and the Defense Mechanisms: A Comprehensive Surveycited by 17
  13. 2023ParaFuzz: An Interpretability-Driven Technique for Detecting Poisoned Samples in NLPcited by 16
  14. 2023Setting the Trap: Capturing and Defeating Backdoors in Pretrained Language Models through Honeypotscited by 16
  15. 2024Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuningcited by 16
  16. 2024Learning to Poison Large Language Models for Downstream Manipulationcited by 15
  17. 2023Large Language Models Are Better Adversaries: Exploring Generative Clean-Label Backdoor Attacks Against Text Classifierscited by 14
  18. 2023TextGuard: Provable Defense against Backdoor Attacks on Text Classificationcited by 14
  19. 2024Mitigating Backdoor Threats to Large Language Models: Advancement and Challengescited by 14
  20. 2022A Survey on Backdoor Attack and Defense in Natural Language Processingcited by 13
  21. 2023BDMMT: Backdoor Sample Detection for Language Models through Model Mutation Testingcited by 13
  22. 2025BadJudge: Backdoor Vulnerabilities of LLM-as-a-Judgecited by 13
  23. 2022WeDef: Weakly Supervised Backdoor Defense for Text Classificationcited by 12
  24. 2024Simulate and Eliminate: Revoke Backdoors for Generative Large Language Modelscited by 11
  25. 2024A Study of Backdoors in Instruction Fine-tuned Language Modelscited by 11
  26. 2023Backdoor Attacks with Input-unique Triggers in NLPcited by 10
  27. 2022Poison Attack and Defense on Deep Source Code Processing Modelscited by 8
  28. 2023Unlearning with Fisher Maskingcited by 8
  29. 2025Efficient Input-level Backdoor Defense on Text-to-Image Synthesis via Neuron Activation Variationcited by 8
  30. 2025A Systematic Review of Poisoning Attacks Against Large Language Modelscited by 8
  31. 2023NCL: Textual Backdoor Defense Using Noise-augmented Contrastive Learningcited by 7
  32. 2023Diffusion Theory as a Scalpel: Detecting and Purifying Poisonous Dimensions in Pre-trained Language Models Caused by Backdoor or Biascited by 7
  33. 2024Acquiring Clean Language Models from Backdoor Poisoned Datasets by Downscaling Frequency Spacecited by 6
  34. 2024Two Heads are Better than One: Nested PoE for Robust Defense Against Multi-Backdoorscited by 6
  35. 2024Rethinking Backdoor Detection Evaluation for Language Modelscited by 6
  36. 2024Trustworthy Text-to-Image Diffusion Models: A Timely and Focused Surveycited by 6
  37. 2026Purifying Generative LLMs from Backdoors without Prior Knowledge or Clean Referencecited by 6
  38. 2024Imperceptible Rhythm Backdoor Attacks: Exploring Rhythm Transformation for Embedding Undetectable Vulnerabilities on Speech Recognitioncited by 5

50 most cited of 80 All 80 (JSON)