SecureLLM: Using Compositionality to Build Provably Secure Language Models for Private, Sensitive, and Secret Data
Abstract: Traditional security mechanisms isolate resources from users who should not access them. We reflect the compositional nature of such security mechanisms back into the structure of LLMs to build a provably secure LLM; that we term SecureLLM. Other approaches to LLM safety attempt to protect against bad actors or bad outcomes, but can only do so to an extent making them inappropriate for sensitive data. SecureLLM blends access security with fine-tuning methods. Each data silo has associated with it a separate fine-tuning and a user has access only to the collection of fine-tunings that they have permission for. The model must then perform on compositional tasks at the intersection of those data silos with the combination of those individual fine-tunings. While applicable to any task like document QA or making API calls, in this work we concern ourselves with models that learn the layouts of new SQL databases to provide natural-language-to-SQL translation capabilities. Existing fine-tuning composition methods fail in this challenging environment, as they are not well-equipped for handling compositional tasks. Compositionality remains a challenge for LLMs. We contribute both a difficult new compositional natural-language-to-SQL translation task and a new perspective on LLM security that allows models to be deployed to secure environments today.
- Deep learning with differential privacy. In Proceedings of the 2016 ACM SIGSAC conference on computer and communications security, pages 308–318.
- The secret sharer: Evaluating and testing unintended memorization in neural networks. In 28th USENIX Security Symposium (USENIX Security 19), pages 267–284.
- Extracting training data from large language models. In 30th USENIX Security Symposium (USENIX Security 21), pages 2633–2650.
- Adaptersoup: Weight averaging to improve generalization of pretrained language models. arXiv preprint arXiv:2302.07027.
- Temporal Data and the Relational Model A detailed investigation into the application of interval and relation theory to the problem of temporal database management. Morgan Kaufmann Publishers.
- Compositional visual generation and inference with energy based models. arXiv preprint arXiv:2004.06030.
- Membership inference attacks on sequence-to-sequence models: Is my data in your machine translation system? Transactions of the Association for Computational Linguistics, 8:49–63.
- Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685.
- Membership inference attacks on machine learning: A survey. ACM Computing Surveys (CSUR), 54(11s):1–37.
- Lorahub: Efficient cross-task generalization via dynamic lora composition. arXiv preprint arXiv:2307.13269.
- Training data leakage analysis in language models. arXiv preprint arXiv:2101.05405.
- Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
- Large language models can be strong differentially private learners. arXiv preprint arXiv:2110.05679.
- Peft: State-of-the-art parameter-efficient fine-tuning methods. https://github.com/huggingface/peft.
- Learning differentially private recurrent language models. arXiv preprint arXiv:1710.06963.
- Comprehensive privacy analysis of deep learning: Passive and active white-box inference attacks against centralized and federated learning. In 2019 IEEE symposium on security and privacy (SP), pages 739–753. IEEE.
- Dropout: a simple way to prevent neural networks from overfitting. The journal of machine learning research, 15(1):1929–1958.
- Multitask pre-training of modular prompt for Chinese few-shot learning. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11156–11172, Toronto, Canada. Association for Computational Linguistics.
- Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288.
- Differentially private fine-tuning of language models. arXiv preprint arXiv:2110.06500.
- Analyzing information leakage of updates to natural language models. In Proceedings of the 2020 ACM SIGSAC conference on computer and communications security, pages 363–375.
- Composing parameter-efficient modules with arithmetic operations. arXiv preprint arXiv:2306.14870.
- On the editing distance between undirected acyclic graphs. International Journal of Foundations of Computer Science, 7(01):43–57.
- Provably confidential language modelling. arXiv preprint arXiv:2205.01863.
- Don’t forget private retrieval: distributed private similarity search for large language models. arXiv preprint arXiv:2311.12955.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.