EMNLP Findings2020
TextHide: Tackling Data Privacy for Language Understanding Tasks
Yangsibo Huang, Zhao Song, Danqi Chen, Kai Li, Sanjeev Arora
Abstract
An unsolved challenge in distributed or federated learning is to effectively mitigate privacy risks without slowing down training or reducing accuracy. In this paper, we propose Tex-tHide aiming at addressing this challenge for natural language understanding tasks. It requires all participants to add a simple encryption step to prevent an eavesdropping attacker from recovering private text data. Such an encryption step is efficient and only affects the task performance slightly. In addition, Tex-tHide fits well with the popular framework of fine-tuning pre-trained language models (e.g., BERT) for any sentence or sentence-pair task. We evaluate TextHide on the GLUE benchmark, and our experiments show that TextHide can effectively defend attacks on shared gradients or representations and the averaged accuracy reduction is only 1.9%. We also present an analysis of the security of TextHide using a conjecture about the computational intractability of a mathematical problem. 1
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
3 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Practical Secure Aggregation for Privacy-Preserving Machine Learning cites Deep Learning with Differential Privacy
- Sentence-level Privacy for Document Embeddings cites Privacy Risks of General-Purpose Language Models
- Sentence-level Privacy for Document Embeddings cites Deep Learning with Differential Privacy
- Sentence-level Privacy for Document Embeddings cites Information Leakage in Embedding Models
- Text Embedding Inversion Security for Multilingual Language Models cites Information Leakage in Embedding Models
- Privacy Risks of General-Purpose Language Models cites Deep Learning with Differential Privacy
- Information Leakage in Embedding Models cites Privacy Risks of General-Purpose Language Models
- Information Leakage in Embedding Models cites Deep Learning with Differential Privacy
- SecureML: A System for Scalable Privacy-Preserving Machine Learning cites Deep Learning with Differential Privacy
Cited by top-tier papers3
- Sentence-level Privacy for Document EmbeddingsCasey Meehan, Khalil Mrini, Kamalika ChaudhuriACL 2022 · 26 citations
- Text Embedding Inversion Security for Multilingual Language ModelsYiyi Chen, Heather C. Lent, Johannes BjervaACL 2024 · 10 citations
- Anonymisation Models for Text Data: State of the art, Challenges and Future DirectionsPierre Lison, Ildikó Pilán, David Sánchez, Montserrat Batet et al.ACL 2021
Builds on6
- Deep Learning with Differential PrivacyMartín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan et al.CCS 2016 · 7,620 citations
- Practical Secure Aggregation for Privacy-Preserving Machine LearningKallista A. Bonawitz, Vladimir Ivanov, Ben Kreuter, Antonio Marcedone et al.CCS 2017 · 3,936 citations
- SecureML: A System for Scalable Privacy-Preserving Machine LearningPayman Mohassel, Yupeng ZhangS&P 2017 · 2,107 citations
- Privacy Risks of General-Purpose Language ModelsXudong Pan, Mi Zhang, Shouling Ji, Min YangS&P 2020 · 291 citations
- Information Leakage in Embedding ModelsCongzheng Song, Ananth RaghunathanCCS 2020 · 200 citations
Related papers
- Federated Model Decomposition with Private Vocabulary for Text ClassificationZhuo Zhang, Xiangjing Hu, Lizhen Qu, Qifan Wang et al.EMNLP 2022 · 5 citations
- TextFusion: Privacy-Preserving Pre-trained Model Inference via Token FusionXin Zhou, Jinzhu Lu, Tao Gui, Ruotian Ma et al.EMNLP 2022 · 12 citations
- InstaHide: Instance-hiding Schemes for Private Distributed LearningYangsibo Huang, Zhao Song, Kai Li, Sanjeev AroraICML 2020 · 178 citations
- Recovering Private Text in Federated Learning of Language ModelsSamyak Gupta, Yangsibo Huang, Zexuan Zhong, Tianyu Gao et al.NeurIPS 2022 · 120 citations
- Empirical Studies of Institutional Federated Learning For Natural Language ProcessingXinghua Zhu, Jianzong Wang, Zhenhou Hong, Jing XiaoEMNLP Findings 2020 · 56 citations
