Provably Confidential Language Modelling
Large language models are shown to memorize privacy information such as social security numbers in training data. Given the sheer scale of the training corpus, it is challenging to screen and filter these privacy data, either manually or automatically. In this paper, we propose Confidentially Redacted Training (CRT), a method to train language generation models while protecting the confidential segments. We borrow ideas from differential privacy (which solves a related but distinct problem) and show that our method is able to provably prevent unintended memorization by randomizing parts of the training process. Moreover, we show that redaction with an approximately correct screening policy amplifies the confidentiality guarantee. We implement the method for both LSTM and GPT language models. Our experimental results show that the models trained by CRT obtain almost the same perplexity while preserving strong confidentiality.
Code (0)
등록된 구현이 없습니다.
Tasks
Language ModellingMemorizationText GenerationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Provably Confidential Language Modelling
Large language models are shown to memorize privacy information such as social security numbers in training data. Given the sheer scale of the training corpus, it is challenging to screen and filter these privacy data, e…
Language ModellingMemorizationText GenerationCaPC Learning: Confidential and Private Collaborative Learning
Machine learning benefits from large training datasets, which may not always be possible to collect by any single entity, especially when using privacy-sensitive data. In many contexts, such as healthcare and finance, se…
FairnessFederated LearningModelling with Discretized Variables
This paper deals with econometric models in which the dependent variable, some explanatory variables, or both are observed as censored interval data. This discretization often happens due to confidentiality of sensitive …
regressionProvably Protecting Fine-Tuned LLMs from Training Data Extraction while Preserving Utility
Fine-tuning large language models (LLMs) on sensitive datasets raises privacy concerns, as training data extraction (TDE) attacks can expose highly confidential information. Existing defenses against such attacks either …
Provably Secure Retrieval-Augmented Generation
Although Retrieval-Augmented Generation (RAG) systems have been widely applied, the privacy and security risks they face, such as data leakage and data poisoning, have not been systematically addressed yet. Existing defe…