VaultGemma: A Differentially Private Gemma Model
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917035315822592 |
|---|---|
| author | Sinha, Amer Mesnard, Thomas McKenna, Ryan Liu, Daogao Choquette-Choo, Christopher A. Huang, Yangsibo Yu, Da Kaissis, George Charles, Zachary Liu, Ruibo Chua, Lynn Kamath, Pritish Manurangsi, Pasin He, Steve Zhang, Chiyuan Ghazi, Badih Pigem, Borja De Balle Eruvbetine, Prem Warkentin, Tris Joulin, Armand Kumar, Ravi |
| author_facet | Sinha, Amer Mesnard, Thomas McKenna, Ryan Liu, Daogao Choquette-Choo, Christopher A. Huang, Yangsibo Yu, Da Kaissis, George Charles, Zachary Liu, Ruibo Chua, Lynn Kamath, Pritish Manurangsi, Pasin He, Steve Zhang, Chiyuan Ghazi, Badih Pigem, Borja De Balle Eruvbetine, Prem Warkentin, Tris Joulin, Armand Kumar, Ravi |
| contents | We introduce VaultGemma 1B, a 1 billion parameter model within the Gemma family, fully trained with differential privacy. Pretrained on the identical data mixture used for the Gemma 2 series, VaultGemma 1B represents a significant step forward in privacy-preserving large language models. We openly release this model to the community |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_15001 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VaultGemma: A Differentially Private Gemma Model Sinha, Amer Mesnard, Thomas McKenna, Ryan Liu, Daogao Choquette-Choo, Christopher A. Huang, Yangsibo Yu, Da Kaissis, George Charles, Zachary Liu, Ruibo Chua, Lynn Kamath, Pritish Manurangsi, Pasin He, Steve Zhang, Chiyuan Ghazi, Badih Pigem, Borja De Balle Eruvbetine, Prem Warkentin, Tris Joulin, Armand Kumar, Ravi Cryptography and Security Artificial Intelligence We introduce VaultGemma 1B, a 1 billion parameter model within the Gemma family, fully trained with differential privacy. Pretrained on the identical data mixture used for the Gemma 2 series, VaultGemma 1B represents a significant step forward in privacy-preserving large language models. We openly release this model to the community |
| title | VaultGemma: A Differentially Private Gemma Model |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2510.15001 |