5 papers
The Illusion of Certainty: Uncertainty Quantification for LLMs Fails under Ambiguity
Tim Tomov, Dominik Fuchsgruber, Tom Wollschläger +1
Accurate uncertainty quantification (UQ) in Large Language Models (LLMs) is critical for trustworthy deployment. While real-world language is inherently ambiguous, reflecting aleat…
Diffusion LLMs are Natural Adversaries for any LLM
David Lüdke, Tom Wollschläger, Paul Ungermann +2
We introduce a novel framework that transforms the resource-intensive (adversarial) prompt optimization problem into an \emph{efficient, amortized inference task}. Our core insight…
REINFORCE Adversarial Attacks on Large Language Models: An Adaptive, Distributional, and Semantic Objective
Simon Geisler, Tom Wollschläger, M. H. I. Abdalla +3
To circumvent the alignment of large language models (LLMs), current optimization-based adversarial attacks usually craft adversarial prompts by maximizing the likelihood of a so-c…
Adversarial Alignment for LLMs Requires Simpler, Reproducible, and More Measurable Objectives
Leo Schwinn, Yan Scholten, Tom Wollschläger +4
Misaligned research objectives have considerably hindered progress in adversarial robustness research over the past decade. For instance, an extensive focus on optimizing target me…
The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence
Tom Wollschläger, Jannes Elstner, Simon Geisler +3
The safety alignment of large language models (LLMs) can be circumvented through adversarially crafted inputs, yet the mechanisms by which these attacks bypass safety barriers rema…