mlsanityData to 14 Oct 2026

Text Revealer: Private Text Reconstruction via Model Inversion Attacks against Transformers

arXiv 2209.10505 ↗First posted 21 Sep 2022cs.CLcited by 31

Not on the map. Why?

Abstract (arXiv)

Text classification has become widely used in various natural language processing applications like sentiment analysis. Current applications often use large transformer-based language models to classify input texts. However, there is a lack of systematic study on how much private information can be inverted when publishing models. In this paper, we formulate Text Revealer -- the first model inversion attack for text reconstruction against text classification with transformers. Our attacks faithfully reconstruct private texts included in training data with access to the target model. We leverage an external dataset and GPT-2 to generate the target domain-like fluent text, and then perturb its hidden state optimally with the feedback from the target model. Our extensive experiments demonstrate that our attacks are effective for datasets with different text lengths and can reconstruct private texts with accuracy.

Key facts. Cited by 31 papers in the map (data through 14 Oct 2026). Builds on 18 of its 29 references that are in the map. First posted on arXiv 21 Sep 2022.

Builds on

18 of 29 references.

References · 18

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  3. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  4. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  5. 2020Extracting Training Data from Large Language Modelscited by 2,189
  6. 2019TinyBERT: Distilling BERT for Natural Language Understandingcited by 1,167
  7. 2019Deep Leakage from Gradientscited by 1,021
  8. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 799
  9. 2020Systematic Evaluation of Privacy Risks of Machine Learning Modelscited by 292
  10. 2019The Secret Revealer: Generative Model-Inversion Attacks Against Deep Neural Networkscited by 246
  11. 2021Does BERT Pretrained on Clinical Notes Reveal Sensitive Data?cited by 93
  12. 2022Variational Model Inversion Attackscited by 75
6 more
  1. 2022BiBERT: Accurate Fully Binarized BERTcited by 73
  2. 2021Generalization in NLI: Ways (Not) To Go Beyond Simple Heuristicscited by 67
  3. 2021TAG: Gradient Attack on Transformer-based Language Modelscited by 63
  4. 2022Label-Only Model Inversion Attacks via Boundary Repulsioncited by 56
  5. 2019Adversarial Neural Network Inversion via Auxiliary Knowledge Alignmentcited by 32
  6. 2021Elbert: Fast Albert with Confidence-Window Based Early Exitcited by 14

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 8; 2024: 10; 2025: 7; 2026: 6.

31 papers.

Sort:

By area:

  1. 2023A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Uglycited by 387
  2. 2023Analyzing Leakage of Personally Identifiable Information in Language Modelscited by 285
  3. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deploymentcited by 116
  4. 2023Language Model Inversioncited by 68
  5. 2023CodeLMSec Benchmark: Systematically Evaluating and Finding Security Vulnerabilities in Black-Box Code Language Modelscited by 55
  6. 2023Privacy in Large Language Models: Attacks, Defenses and Future Directionscited by 51
  7. 2023Training Data Extraction From Pre-trained Language Models: A Surveycited by 45
  8. 2024Feedback-Guided Extraction of Knowledge Base from Retrieval-Augmented LLM Applicationscited by 37
  9. 2024On the Privacy Risk of In-context Learningcited by 32
  10. 2023Identifying and Mitigating Privacy Risks Stemming from Language Models: A Surveycited by 30
  11. 2024Privacy Preserving Prompt Engineering: A Surveycited by 25
  12. 2025A Survey on Privacy Risks and Protection in Large Language Modelscited by 21
19 more
  1. 2025A Survey on Model Extraction Attacks and Defenses for Large Language Modelscited by 19
  2. 2023On Strengthening and Defending Graph Reconstruction Attack with Markov Chain Approximationcited by 16
  3. 2024Model Inversion Attacks: A Survey of Approaches and Countermeasurescited by 11
  4. 2025LLM Security: Vulnerabilities, Attacks, Defenses, and Countermeasurescited by 8
  5. 2024DORY: Deliberative Prompt Recovery for LLMcited by 7
  6. 2026Window-based Membership Inference Attacks Against Fine-tuned Large Language Modelscited by 7
  7. 2024Privacy-preserving Fine-tuning of Large Language Models through Flatnesscited by 6
  8. 2024Was it Slander? Towards Exact Inversion of Generative Language Modelscited by 3
  9. 2024Privacy Evaluation Benchmarks for NLP Modelscited by 3
  10. 2024Towards Data Governance of Frontier AI Modelscited by 3
  11. 2025On The Role of Pretrained Language Models in General-Purpose Text Embeddings: A Surveycited by 3
  12. 2025Securing AI Systems: A Guide to Known Attacks and Impactscited by 1
  13. 2026Data-Free Privacy-Preserving for LLMs via Model Inversion and Selective Unlearningcited by 1
  14. 2026Efficient and Privacy Aware Edge Cloud Collaborative Inference for Large Language Modelscited by 1
  15. 2024MRP-LLM: Multitask Reflective Large Language Models for Privacy-Preserving Next POI Recommendationcited by 0
  16. 2025Model Inversion meets Cryptographic Fuzzy Extractorscited by 0
  17. 2026Beyond Homophily: Towards Generalized Graph Reconstruction Attack and Defensecited by 0
  18. 2026Domain Agnostic Text Redaction from Natural Language Rules using Instruction Tuningcited by 0
  19. 2026Inadvertent Context Leakage in Language Modelscited by 0