mlsanityData to 7 Oct 2026

Poison Attacks against Text Datasets with Conditional Adversarially Regularized Autoencoder

arXiv 2010.02684 ↗First posted 6 Oct 2020cs.CL cs.AI cs.NEcited by 27

Not on the map. Why?

Abstract (arXiv)

This paper demonstrates a fatal vulnerability in natural language inference (NLI) and text classification systems. More concretely, we present a 'backdoor poisoning' attack on NLP models. Our poisoning attack utilizes conditional adversarially regularized autoencoder (CARA) to generate poisoned training samples by poison injection in latent space. Just by adding 1% poisoned data, our experiments show that a victim BERT finetuned classifier's predictions can be steered to the poison target class with success rates of >80% when the input hypothesis is injected with the poison signature, demonstrating that NLI and text classification systems face a huge security risk.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Oct 2020.

Builds on

No reference list yet.

Built on by

Cited by year

Citing papers per year: 2021: 6; 2022: 9; 2023: 7; 2024: 4; 2025: 1.

27 papers.

Sort:

By area:

  1. 2023Poisoning Language Models During Instruction Tuningcited by 254
  2. 2022Impact of Pretraining Term Frequencies on Few-Shot Reasoningcited by 142
  3. 2024What is in Your Safe Data? Identifying Benign Data that Breaks Safetycited by 114
  4. 2021RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Modelscited by 97
  5. 2021Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Modelscited by 96
  6. 2022A Study of the Attention Abnormality in Trojaned BERTscited by 47
  7. 2022BITE: Textual Backdoor Attacks with Iterative Trigger Injectioncited by 47
  8. 2023TrojanPuzzle: Covertly Poisoning Code-Suggestion Modelscited by 47
  9. 2021Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-Level Backdoor Attackscited by 44
  10. 2021CoProtector: Protect Open-Source Code against Unauthorized Training Usage with Data Poisoningcited by 39
  11. 2021Revisiting Methods for Finding Influential Examplescited by 39
  12. 2022Rethinking the Reverse-engineering of Trojan Triggerscited by 37
15 more
  1. 2023Did You Train on My Dataset? Towards Public Dataset Protection with Clean-Label Backdoor Watermarkingcited by 37
  2. 2024PoisonBench: Assessing Large Language Model Vulnerability to Data Poisoningcited by 31
  3. 2022Kallima: A Clean-label Framework for Textual Backdoor Attackscited by 22
  4. 2022Rank List Sensitivity of Recommender Systems to Interaction Perturbationscited by 20
  5. 2023Attacks in Adversarial Machine Learning: A Systematic Survey from the Life-cycle Perspectivecited by 19
  6. 2023Large Language Models Are Better Adversaries: Exploring Generative Clean-Label Backdoor Attacks Against Text Classifierscited by 14
  7. 2023BDMMT: Backdoor Sample Detection for Language Models through Model Mutation Testingcited by 13
  8. 2022A Survey on AI Sustainability: Emerging Trends on Learning Algorithms and Research Challengescited by 7
  9. 2022Attention Hijacking in Trojan Transformerscited by 6
  10. 2024SpamDam: Towards Privacy-Preserving and Adversary-Resistant SMS Spam Detectioncited by 3
  11. 2025The Ultimate Cookbook for Invisible Poison: Crafting Subtle Clean-Label Text Backdoors with Style Attributescited by 2
  12. 2021Classification Auto-Encoder based Detector against Diverse Data Poisoning Attackscited by 1
  13. 2022Detecting Backdoors in Deep Text Classifierscited by 0
  14. 2023Context Matters: Data-Efficient Augmentation of Large Language Models for Scientific Applicationscited by 0
  15. 2024Manipulating Predictions over Discrete Inputs in Machine Teachingcited by 0