Muhammad Huzaifa, Lea Schönherr, and Thorsten Eisenhofer

WISE-ATTA: When to Ask for Labels in Budgeted Active Test-Time Adaptation

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) pdf code arxiv

Sina Mavali, David Pape, Jonathan Evertz, Samira Abedini, Devansh Srivastav, Thorsten Eisenhofer, Sahar Abdelnabi, and Lea Schönherr

No More, No Less: Task Alignment in Terminal Agents

NeurIPS Evaluations and Datasets Track pdf arxiv

Jonathan Evertz, Niklas Risse, Nicolai Neuer, Andreas Müller, Philipp Normann, Gaetano Sapia, Srishti Gupta, David Pape, Soumya Shaw, Devansh Srivastav, Christian Wressnegger, Erwin Quiring, Thorsten Eisenhofer, Daniel Arp, and Lea Schönherr

Chasing Shadows: Pitfalls in LLM Security Research

Network and Distributed System Security Symposium (NDSS) pdf website code arxiv

Felix Weissberg, Lukas Pirch, Erik Imgrund, Jonas Möller, Thorsten Eisenhofer, and Konrad Rieck

LLM-based Vulnerability Discovery through the Lens of Code Metrics

IEEE/ACM International Conference on Software Engineering (ICSE) pdf code

Jonathan Evertz, Merlin Chlosta, Lea Schönherr, and Thorsten Eisenhofer

Whispers in the Machine: Confidentiality in Agentic Systems

Detection of Intrusions and Malware & Vulnerability Assessment (DIMVA) pdf code arxiv

Mohammad Ebrahimi Fard, Felix Weissberg, Erik Imgrund, Thorsten Eisenhofer, and Konrad Rieck

Shape-Shifting Malicious Code in Software Backdoors via Language Models

ACM Asia Conference on Computer and Communications Security (ASIACCS) pdf code

Pia Hanfeld, Erik Imgrund, Felix Weissberg, Thorsten Eisenhofer, Wolfgang Hönig, and Konrad Rieck

Controlling Autonomous Vehicles using Continuous Adversarial Patches

ACM Workshop on Artificial Intelligence and Security (AISec) pdf code Spotlight Paper

Lennart von Stülpnagel, Felix Weissberg, Jonas Möller, Benjamin Greb, Thorsten Eisenhofer, and Konrad Rieck

Data Augmentation in Learning-Based Binary Code Analysis

ACM Workshop on Artificial Intelligence and Security (AISec) pdf code data

Muhammad Huzaifa, Sina Mavali, and Thorsten Eisenhofer

Safety of Latent Communication in Multi-Agent Systems

Computing Research Repository (CoRR) pdf arxiv

Jonas Möller, Erik Imgrund, Thorsten Eisenhofer, and Konrad Rieck

Hardware-Triggered Backdoors

Computing Research Repository (CoRR) pdf code arxiv

Erik Imgrund, Thorsten Eisenhofer, and Konrad Rieck

Adversarial Observations in Weather Forecasting

ACM Conference on Computer and Communications Security (CCS) pdf code poster arxiv Distinguished Paper Award

Jonas Möller, Lukas Pirch, Felix Weissberg, Sebastian Baunsgaard, Thorsten Eisenhofer, and Konrad Rieck

Adversarial Inputs for Linear Algebra Backends

International Conference on Machine Learning (ICML) pdf code

Felix Weissberg, Jan Malte Hilgefort, Steve Grogorick, Daniel Arp, Thorsten Eisenhofer, Martin Eisemann, and Konrad Rieck

Seeing Through: Analyzing and Attacking Virtual Backgrounds in Video Calls

USENIX Security Symposium pdf code poster

David Pape, Sina Mavali, Thorsten Eisenhofer, and Lea Schönherr

Prompt Obfuscation for Large Language Models

USENIX Security Symposium pdf code arxiv

Thorsten Eisenhofer, Doreen Riepel, Varun Chandrasekaran, Esha Ghosh, Olga Ohrimenko, and Nicolas Papernot

Verifiable and Provably Secure Machine Unlearning

IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) pdf code slides poster arxiv

David Beste, Grégoire Menguy, Hossein Hajipour, Mario Fritz, Antonio Emanuele Cinà, Sébastien Bardin, Thorsten Holz, Thorsten Eisenhofer, and Lea Schönherr

Exploring the Potential of LLMs for Code Deobfuscation

Detection of Intrusions and Malware & Vulnerability Assessment (DIMVA) pdf code

Roei Schuster, Jin Peng Zhou, Thorsten Eisenhofer, Paul Grubbs, and Nicolas Papernot

Learned-Database Systems Security

Transactions on Machine Learning Research (TMLR) pdf arxiv

Joel Frank, Franziska Herbert, Jonas Ricker, Lea Schönherr, Thorsten Eisenhofer, Asja Fischer, Markus Dürmuth, and Thorsten Holz

A Representative Study on Human Detection of Artificially Generated Media Across Countries

IEEE Symposium on Security and Privacy (S&P) pdf preregistration code arxiv

Felix Weissberg, Jonas Möller, Tom Ganz, Erik Imgrund, Lukas Pirch, Lukas Seidel, Moritz Schloegel, Thorsten Eisenhofer, and Konrad Rieck

SoK: Where to Fuzz? Assessing Target Selection Methods in Directed Fuzzing

ACM Asia Conference on Computer and Communications Security (ASIACCS) pdf code arxiv

Jonas Möller, Felix Weissberg, Lukas Pirch, Thorsten Eisenhofer, and Konrad Rieck

Cross-Language Differential Testing of JSON Parsers

ACM Asia Conference on Computer and Communications Security (ASIACCS) pdf code

Thorsten Eisenhofer

Security of Machine Learning Systems

Dissertation pdf slides Award for Outstanding Doctoral Achievement

Thorsten Eisenhofer, Erwin Quiring, Jonas Möller, Doreen Riepel, Thorsten Holz, and Konrad Rieck

No more Reviewer #2: Subverting Automatic Paper-Reviewer Assignment using Adversarial Learning

USENIX Security Symposium pdf slides examples code arxiv

Hojjat Aghakhani, Lea Schönherr, Thorsten Eisenhofer, Dorothea Kolossa, Thorsten Holz, Christopher Kruegel, and Giovanni Vigna

VenoMave: Targeted Poisoning Against Speech Recognition

IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) pdf code arxiv

Nico Schiller, Merlin Chlosta, Moritz Schloegel, Nils Bars, Thorsten Eisenhofer, Tobias Scharnowski, Felix Domke, Lea Schönherr, and Thorsten Holz

Drone Security and the Mysterious Case of DJI's DroneID

Network and Distributed System Security Symposium (NDSS) pdf code

David Pape, Sina Däubener, Thorsten Eisenhofer, Antonio Emanuele Cinà, and Lea Schönherr

On the Limitations of Model Stealing with Uncertainty Quantification Models

European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN) pdf arxiv

Michel Abdalla, Thorsten Eisenhofer, Eike Kiltz, Sabrina Kunzweiler, and Doreen Riepel

Password-Authenticated Key Exchange from Group Actions

Annual International Cryptology Conference (CRYPTO) pdf

Lea Schönherr, Maximilian Golla, Thorsten Eisenhofer, Jan Wiele, Dorothea Kolossa, and Thorsten Holz

Exploring Accidental Triggers of Smart Speakers

Computer Speech & Language (CSL) pdf website arxiv

Thorsten Eisenhofer, Lea Schönherr, Joel Frank, Lars Speckemeier, Dorothea Kolossa, and Thorsten Holz

Dompteur: Taming Audio Adversarial Examples

USENIX Security Symposium pdf talk slides code arxiv

Joel Frank, Thorsten Eisenhofer, Lea Schönherr, Asja Fischer, Dorothea Kolossa, and Thorsten Holz

Leveraging Frequency Analysis for Deep Fake Image Recognition

International Conference on Machine Learning (ICML) pdf slides code arxiv

Lea Schönherr, Thorsten Eisenhofer, Steffen Zeiler, Thorsten Holz, and Dorothea Kolossa

Imperio: Robust Over-the-Air Adversarial Examples for Automatic Speech Recognition Systems

Annual Computer Security Applications Conference (ACSAC) pdf talk examples arxiv