E-mail senden E-Mail Adresse kopieren

E-Mail

Adresse

Im Oberen Werk 1
66386 St. Ingbert (Germany)

Awards (Auswahl)

2025: Werner-von-Siemens-Fellow

2025: ERC Starting Grant

2024: GI Junior-Fellow 

2024: Busy Beaver Award "Differential Privacy: Mathematical Foundations and Applications in Machine Learning“, Saarland University

 

Weitere Informationen

Kurzbiografie

Franziska Boenisch ist Tenure-Track Faculty am CISPA Helmholtz-Zentrum für Informationssicherheit. Am CISPA ist sie Co-Leiterin des SprintML Lab (Secure, Private, Robust, Interpretable, and Trustworthy Machine Learning), in dem sie zur Weiterentwicklung von vertrauenswürdigem Machine Learning forscht. Zuvor war sie Postdoctoral Fellow am Vector Institute for Artificial Intelligence, betreut von Prof. Dr. Nicolas Papernot. Vor ihrem Wechsel an das Vector Institute war sie Doktorandin an der Freien Universität Berlin sowie wissenschaftliche Mitarbeiterin am Fraunhofer-Institut für Angewandte und Integrierte Sicherheit (AISEC).

CV: Letzte Stationen

Seit 2023
Tenure-Track Faculty am CISPA
2022 - 2023
Postdoctoral Fellow - Vector Institute for Artificial Intelligence, Toronto
2019 - 2022
PhD Student und Research Associate - Department of Secure Systems Engineering, Fraunhofer AISEC

Veröffentlichungen von Franziska Boenisch

Jahr 2024

Konferenz / Medium

ICML Workshop on Foundation Models in the WIld (ICML-W) POST: A Framework for Privacy of Soft-prompt Transfer

Konferenz / Medium

IEEE International Symposium on Information Theory (ISIT) Controlled privacy leakage propagation throughout differential private overlapping grouped learning

Artikel

IEEE Journal on Selected Areas in Information Theory Controlled privacy leakage propagation throughout overlapping grouped learning

Konferenz / Medium

International Conference on Learning Representations (ICLR) Memorization in Self-Supervised Learning Improves Downstream Generalization

Konferenz / Medium

Conference on Neural Information Processing Systems (NeurIPS) Finding NeMo: Localizing Neurons Responsible For Memorization in Diffusion Models

Konferenz / Medium

Conference on Neural Information Processing Systems (NeurIPS) Localizing Memorization in SSL Vision Encoders

Konferenz / Medium

Conference on Neural Information Processing Systems (NeurIPS) Open LLMs are Necessary for Private Adaptations and Outperform their Closed Alternatives

Jahr 2023

Konferenz / Medium

Conference on Neural Information Processing Systems (NeurIPS) Bucks for Buckets (B4B): Active Defenses Against Stealing Encoders

Konferenz / Medium

Conference on Neural Information Processing Systems (NeurIPS) Flocks of Stochastic Parrots: Differentially Private Prompt Learning for Large Language Models

Konferenz / Medium

NeurIPS-Workshop (NeurIPS-W)