Igamberdiev, Timour ; Arnold, Thomas ; Habernal, Ivan (2022)
DP-Rewrite: Towards Reproducibility and Transparency in Differentially Private Text Rewriting.
29th International Conference on Computational Linguistics (COLING 2022). Gyeongju, Republic of Korea (12.10.2022-17.10.2022)
Konferenzveröffentlichung, Bibliographie
Kurzbeschreibung (Abstract)
Text rewriting with differential privacy (DP) provides concrete theoretical guarantees for protecting the privacy of individuals in textual documents. In practice, existing systems may lack the means to validate their privacy-preserving claims, leading to problems of transparency and reproducibility. We introduce DP-Rewrite, an open-source framework for differentially private text rewriting which aims to solve these problems by being modular, extensible, and highly customizable. Our system incorporates a variety of downstream datasets, models, pre-training procedures, and evaluation metrics to provide a flexible way to lead and validate private text rewriting research. To demonstrate our software in practice, we provide a set of experiments as a case study on the ADePT DP text rewriting system, detecting a privacy leak in its pre-training approach. Our system is publicly available, and we hope that it will help the community to make DP text rewriting research more accessible and transparent.
| Typ des Eintrags: | Konferenzveröffentlichung |
|---|---|
| Erschienen: | 2022 |
| Autor(en): | Igamberdiev, Timour ; Arnold, Thomas ; Habernal, Ivan |
| Art des Eintrags: | Bibliographie |
| Titel: | DP-Rewrite: Towards Reproducibility and Transparency in Differentially Private Text Rewriting |
| Sprache: | Englisch |
| Publikationsjahr: | 19 Oktober 2022 |
| Verlag: | International Committee on Computational Linguistics |
| Buchtitel: | Proceedings of the Main Conference: The 29th International Conference on Computational Linguistics |
| Reihe: | International Conference on Computational Linguistics: Proceedings of the Conference and Workshops |
| Band einer Reihe: | 29 |
| Veranstaltungstitel: | 29th International Conference on Computational Linguistics (COLING 2022) |
| Veranstaltungsort: | Gyeongju, Republic of Korea |
| Veranstaltungsdatum: | 12.10.2022-17.10.2022 |
| URL / URN: | https://aclanthology.org/2022.coling-1.258 |
| Kurzbeschreibung (Abstract): | Text rewriting with differential privacy (DP) provides concrete theoretical guarantees for protecting the privacy of individuals in textual documents. In practice, existing systems may lack the means to validate their privacy-preserving claims, leading to problems of transparency and reproducibility. We introduce DP-Rewrite, an open-source framework for differentially private text rewriting which aims to solve these problems by being modular, extensible, and highly customizable. Our system incorporates a variety of downstream datasets, models, pre-training procedures, and evaluation metrics to provide a flexible way to lead and validate private text rewriting research. To demonstrate our software in practice, we provide a set of experiments as a case study on the ADePT DP text rewriting system, detecting a privacy leak in its pre-training approach. Our system is publicly available, and we hope that it will help the community to make DP text rewriting research more accessible and transparent. |
| Freie Schlagworte: | UKP_p_crisp_senpai |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Ubiquitäre Wissensverarbeitung |
| Hinterlegungsdatum: | 25 Okt 2022 08:19 |
| Letzte Änderung: | 14 Feb 2023 14:09 |
| PPN: | 505035464 |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum