208 citations · 293 across the 14 of their papers we have counts for
17 papers
WISE-ATTA: When to Ask for Labels in Budgeted Active Test-Time Adaptation
Muhammad Huzaifa, Lea Schönherr, Thorsten Eisenhofer
Active test-time adaptation (ATTA) improves robustness under distribution shift by updating a deployed model during inference while selectively querying supervision. However, most…
No More, No Less: Task Alignment in Terminal Agents
Sina Mavali, David Pape, Jonathan Evertz +5
Terminal agents are increasingly capable of executing complex, long-horizon tasks autonomously from a single user prompt. To do so, they must interpret instructions encountered in…
Unknown Unknowns: Do Hidden Intentions in LLMs Evade Detection?
Devansh Srivastav, David Pape, Lea Schönherr
LLMs expand accessibility and provide wide-reaching access to information. Yet these interactions also create opportunities to embed subtle, goal-oriented behaviours that shape wha…
Chasing Shadows: Pitfalls in LLM Security Research
Jonathan Evertz, Niklas Risse, Nicolai Neuer +12
Large language models (LLMs) are increasingly prevalent in security research. Their unique characteristics, however, introduce challenges that undermine established paradigms of re…
"That's another doom I haven't thought about": A User Study on AI Labels as a Safeguard Against Image-Based Misinformation
Sandra Höltervennhoff, Jonas Ricker, Maike M. Raphael +6
As generative AI is increasingly contributing to the spread of deceptively realistic misinformation, lawmakers have introduced regulations requiring the disclosure of AI-generated…
HexaCoder: Secure Code Generation via Oracle-Guided Synthetic Training Data
Hossein Hajipour, Lea Schönherr, Thorsten Holz +1
Large language models (LLMs) have shown great potential for automatic code generation and form the basis for various tools such as GitHub Copilot. However, recent studies highlight…
Prompt Obfuscation for Large Language Models
David Pape, Sina Mavali, Thorsten Eisenhofer +1
System prompts that include detailed instructions to describe the task performed by the underlying LLM can easily transform foundation models into tools and services with minimal o…
Rag and Roll: An End-to-End Evaluation of Indirect Prompt Manipulations in LLM-based Application Frameworks
Gianluca De Stefano, Lea Schönherr, Giancarlo Pellegrino
Retrieval Augmented Generation (RAG) is a technique commonly used to equip models with out of distribution knowledge. This process involves collecting, indexing, retrieving, and pr…
Conning the Crypto Conman: End-to-End Analysis of Cryptocurrency-based Technical Support Scams
Bhupendra Acharya, Muhammad Saad, Antonio Emanuele Cinà +5
The mainstream adoption of cryptocurrencies has led to a surge in wallet-related issues reported by ordinary users on social media platforms. In parallel, there is an increase in a…
A Representative Study on Human Detection of Artificially Generated Media Across Countries
Joel Frank, Franziska Herbert, Jonas Ricker +5
AI-generated media has become a threat to our digital society as we know it. These forgeries can be created automatically and on a large scale based on publicly available technolog…
On the Limitations of Model Stealing with Uncertainty Quantification Models
David Pape, Sina Däubener, Thorsten Eisenhofer +2
Model stealing aims at inferring a victim model's functionality at a fraction of the original training cost. While the goal is clear, in practice the model's architecture, weight d…
WaveFake: A Data Set to Facilitate Audio Deepfake Detection
Joel Frank, Lea Schönherr
Deep generative modeling has the potential to cause significant harm to society. Recognizing this threat, a magnitude of research into detecting so-called "Deepfakes" has emerged.…
Dompteur: Taming Audio Adversarial Examples
Thorsten Eisenhofer, Lea Schönherr, Joel Frank +3
Adversarial examples seem to be inevitable. These specifically crafted inputs allow attackers to arbitrarily manipulate machine learning systems. Even worse, they often seem harmle…
Unacceptable, where is my privacy? Exploring Accidental Triggers of Smart Speakers
Lea Schönherr, Maximilian Golla, Thorsten Eisenhofer +3
Voice assistants like Amazon's Alexa, Google's Assistant, or Apple's Siri, have become the primary (voice) interface in smart speakers that can be found in millions of households.…
Leveraging Frequency Analysis for Deep Fake Image Recognition
Joel Frank, Thorsten Eisenhofer, Lea Schönherr +3
Deep neural networks can generate images that are astonishingly realistic, so much so that it is often hard for humans to distinguish them from actual photos. These achievements ha…
Imperio: Robust Over-the-Air Adversarial Examples for Automatic Speech Recognition Systems
Lea Schönherr, Thorsten Eisenhofer, Steffen Zeiler +2
Automatic speech recognition (ASR) systems can be fooled via targeted adversarial examples, which induce the ASR to produce arbitrary transcriptions in response to altered audio si…
Adversarial Attacks Against Automatic Speech Recognition Systems via Psychoacoustic Hiding
Lea Schönherr, Katharina Kohls, Steffen Zeiler +2
Voice interfaces are becoming accepted widely as input methods for a diverse set of devices. This development is driven by rapid improvements in automatic speech recognition (ASR),…