Nora Hofer, MA
Research Assistant

E-mailnora.hofer@uibk.ac.at
AddressTechnikerstraße 21A, 6020 Innsbruck, Austria
OfficeICT 2W02
Nora Hofer, MA

Research Interests

  • Digital forensics
  • Steganography and steganalysis
  • Machine learning in adversarial environments

Positions

since 07/2021 Research Assistant, Department of Computer Science, University of Innsbruck, Austria
04/2021 – 06/2021 Student Assistant, Department of Computer Science, University of Innsbruck, Austria
2019 – 2021 Software Developer, DeepOpinion Intelligent Automation GmbH, Innsbruck, Austria
2015 – 2019 Technical Staff, SPEED-U-UP GmbH, Innsbruck, Austria

Education

since 10/2020 Supplementary studies in Computer Science, University of Innsbruck, Austria
2018 – 2020 M.A. studies in Management, Communication & IT, Management Center Innsbruck, Austria
2015 – 2018 B.A. studies in Management, Communication & IT, Management Center Innsbruck, Austria

Publications

  • Schlögl, A., Hofer, N., and Böhme, R. Causes and Effects of Unanticipated Numerical Deviations in Neural Network Inference Frameworks. In Conference on Neural Information Processing Systems (NeurIPS). New Orleans, LA, 2023. [PDF] [Publisher] [Video]
  • Hofer, N. and Böhme, R. Progressive JPEGs in the Wild: Implications for Information Hiding and Forensics. In Proceedings of the 11th ACM Workshop on Information Hiding and Multimedia Security (IH&MMSEC). ACM, Chicago, IL, 2023, pp. 47–58. [PDF] [Publisher] [Video]
  • Beneš, M., Hofer, N., and Böhme, R. The Effect of the JPEG Implementation on the Cover–Source Mismatch Error in Image Steganalysis. In Proceedings of the 30th European Signal Processing Conference (EUSIPCO). EURASIP, Belgrade, Serbia, 2022, pp. 1057–1061. [PDF] [Publisher] [Video]
  • Beneš, M., Hofer, N., and Böhme, R. Know Your Library: How the libjpeg Version Influences Compression and Decompression Results. In Proceedings of the 10th ACM Workshop on Information Hiding and Multimedia Security (IH&MMSEC). ACM, 2022, pp. 19–25. [PDF] [Publisher]
  • Hofer, N., Schöttle, P., Rietzler, A., and Stabinger, S. Adversarial Examples Against a BERT ABSA Model – Fooling Bert With L33T, Misspellign, and Punctuation, In International Conference on Availability, Reliability and Security (ARES). 2021. [Publisher] [Video]

Awards

Projects