PAPILLON: Privacy Preservation from Internet-based and Local Language Model Ensembles
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908284674375680 |
|---|---|
| author | Siyan, Li Raghuram, Vethavikashini Chithrra Khattab, Omar Hirschberg, Julia Yu, Zhou |
| author_facet | Siyan, Li Raghuram, Vethavikashini Chithrra Khattab, Omar Hirschberg, Julia Yu, Zhou |
| contents | Users can divulge sensitive information to proprietary LLM providers, raising significant privacy concerns. While open-source models, hosted locally on the user's machine, alleviate some concerns, models that users can host locally are often less capable than proprietary frontier models. Toward preserving user privacy while retaining the best quality, we propose Privacy-Conscious Delegation, a novel task for chaining API-based and local models. We utilize recent public collections of user-LLM interactions to construct a natural benchmark called PUPA, which contains personally identifiable information (PII). To study potential approaches, we devise PAPILLON, a multi-stage LLM pipeline that uses prompt optimization to address a simpler version of our task. Our best pipeline maintains high response quality for 85.5% of user queries while restricting privacy leakage to only 7.5%. We still leave a large margin to the generation quality of proprietary LLMs for future work. Our data and code is available at https://github.com/siyan-sylvia-li/PAPILLON. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_17127 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | PAPILLON: Privacy Preservation from Internet-based and Local Language Model Ensembles Siyan, Li Raghuram, Vethavikashini Chithrra Khattab, Omar Hirschberg, Julia Yu, Zhou Cryptography and Security Computation and Language Users can divulge sensitive information to proprietary LLM providers, raising significant privacy concerns. While open-source models, hosted locally on the user's machine, alleviate some concerns, models that users can host locally are often less capable than proprietary frontier models. Toward preserving user privacy while retaining the best quality, we propose Privacy-Conscious Delegation, a novel task for chaining API-based and local models. We utilize recent public collections of user-LLM interactions to construct a natural benchmark called PUPA, which contains personally identifiable information (PII). To study potential approaches, we devise PAPILLON, a multi-stage LLM pipeline that uses prompt optimization to address a simpler version of our task. Our best pipeline maintains high response quality for 85.5% of user queries while restricting privacy leakage to only 7.5%. We still leave a large margin to the generation quality of proprietary LLMs for future work. Our data and code is available at https://github.com/siyan-sylvia-li/PAPILLON. |
| title | PAPILLON: Privacy Preservation from Internet-based and Local Language Model Ensembles |
| topic | Cryptography and Security Computation and Language |
| url | https://arxiv.org/abs/2410.17127 |