Related papers: The Evaluation Differential: When Frontier AI Mode…
In recent years, the popularity of fingerprint-based biometric authentication systems significantly increased. However, together with many advantages, biometric systems are still vulnerable to presentation attacks (PAs). In particular, this…
Automatic speech recognition (ASR) outcomes serve as input for downstream tasks, substantially impacting the satisfaction level of end-users. Hence, the diagnosis and enhancement of the vulnerabilities present in the ASR model bear…
Most autonomous driving safety benchmarks use time-to-collision (TTC) to assess risk and guide safe behaviour. However, TTC-based methods treat risk as a one-dimensional closing problem, despite the inherently two-dimensional nature of…
Recent work shows that deep neural networks are vulnerable to adversarial examples. Much work studies adversarial example generation, while very little work focuses on more critical adversarial defense. Existing adversarial detection…
A comprehensive approach to addressing catastrophic risks from AI models should cover the full model lifecycle. This paper explores contingency plans for cases where pre-deployment risk management falls short: where either very dangerous…
We study causal discovery from a single observed sequence of discrete events generated by a stochastic process, as encountered in vehicle logs, manufacturing systems, or patient trajectories. This regime is particularly challenging due to…
Autonomous vehicles in interactive traffic environments are often limited by the scarcity of safety-critical tail events in static datasets, which biases learned policies toward average-case behaviors and reduces robustness. Existing…
In economic program evaluation, it is common to obtain panel data in which outcomes are indicators that an individual has reached an absorbing state. For example, they may indicate whether an individual has exited a period of unemployment,…
Neural network-based approaches have recently shown significant promise in solving partial differential equations (PDEs) in science and engineering, especially in scenarios featuring complex domains or incorporation of empirical data. One…
As frontier AI systems advance toward transformative capabilities, we need a parallel transformation in how we measure and evaluate these systems to ensure safety and inform governance. While benchmarks have been the primary method for…
The ever-evolving landscape of attacks, coupled with the growing complexity of ICT systems, makes crafting anomaly-based intrusion detectors (ID) and error detectors (ED) a difficult task: they must accurately detect attacks, and they…
Every AI benchmark operationalizes theoretical assumptions about the capability it claims to assess. When assumptions function as unexamined commitments, benchmarks stabilize the dominant paradigm by narrowing what counts as progress. Over…
Reliability of deep learning models is critical for deployment in high-stakes applications, where out-of-distribution or adversarial inputs may lead to detrimental outcomes. Evidential Deep Learning, an efficient paradigm for uncertainty…
Robustness of machine learning models is critical for security related applications, where real-world adversaries are uniquely focused on evading neural network based detectors. Prior work mainly focus on crafting adversarial examples (AEs)…
Most existing emotion analysis emphasizes which emotion arises (e.g., happy, sad, angry) but neglects the deeper why. We propose Emotion Interpretation (EI), focusing on causal factors-whether explicit (e.g., observable objects,…
The deployment of AI systems in safety-critical domains, such as industrial defect inspection, autonomous driving, and medical diagnosis, is severely hampered by their lack of reliability. A single undetected erroneous prediction can lead…
Recent work has demonstrated the plausibility of frontier AI models scheming -- knowingly and covertly pursuing an objective misaligned with its developer's intentions. Such behavior could be very hard to detect, and if present in future…
Large language models (LLMs) can internally distinguish between evaluation and deployment contexts, a behaviour known as \emph{evaluation awareness}. This undermines AI safety evaluations, as models may conceal dangerous capabilities during…
Now that AI-driven moderation has become pervasive in everyday life, we often hear claims that "the AI is biased". While this is often said jokingly, the light-hearted remark reflects a deeper concern. How can we be certain that an online…
Black-box safety evaluation of AI systems assumes model behavior on test distributions reliably predicts deployment performance. We formalize and challenge this assumption through latent context-conditioned policies -- models whose outputs…