Related papers: The Evaluation Differential: When Frontier AI Mode…
Representational distinctions within categories are important in all perceptual modalities and also in cognitive and motor representations. Recent pattern-information studies of brain activity have used condition-rich designs to sample the…
Current AI-driven educational systems primarily rely on behavioural analytics, performance metrics, and content-level interactions to model learning. While these approaches provide useful indicators of learner activity, they are…
AI governance frameworks increasingly rely on audits, yet the results of their underlying evaluations require interpretation and context to be meaningfully informative. Even technically rigorous evaluations can offer little useful insight…
Large language models (LLMs) can use in-context demonstrations to improve performance on zero-shot tasks. However, selecting the best in-context examples is challenging because model performance can vary widely depending on the selected…
As autonomous AI agents are increasingly deployed in high-stakes environments, ensuring their safety and alignment with human values is becoming a practical deployment concern. Current benchmarks for AI agents primarily evaluate refusal of…
As artificial intelligence (AI) models are scaled up, new capabilities can emerge unintentionally and unpredictably, some of which might be dangerous. In response, dangerous capabilities evaluations have emerged as a new risk assessment…
Despite the enormous performance of deepneural networks (DNNs), recent studies have shown theirvulnerability to adversarial examples (AEs), i.e., care-fully perturbed inputs designed to fool the targetedDNN. Currently, the literature is…
Recent work has suggested post-hoc explainers might be ineffective for detecting spurious correlations in Deep Neural Networks (DNNs). However, we show there are serious weaknesses with the existing evaluation frameworks for this setting.…
Generalizing deep reinforcement learning agents to unseen environments remains a significant challenge. One promising solution is Unsupervised Environment Design (UED), a co-evolutionary framework in which a teacher adaptively generates…
Two key identifying assumptions used to justify difference-in-differences are parallel trends and no anticipation, yet both may fail in practice. I propose a class of assumptions on anticipation and derive closed-form, sharp bounds on the…
Recent AI media detectors report near-perfect performance under clean laboratory evaluation, yet their robustness under realistic deployment conditions remains underexplored. In practice, AI-generated images are resized, compressed,…
Reinforcement learning (RL) in non-stationary environments is challenging, as changing dynamics and rewards quickly make past experiences outdated. Traditional experience replay (ER) methods, especially those using TD-error prioritization,…
AI agents execute complex multi-step processes, but current evaluation falls short: outcome metrics report success or failure without explaining why, and process-level approaches struggle to connect failure types to their precise locations…
Anomaly Detection (AD) defines the task of identifying observations or events that deviate from typical - or normal - patterns, a critical capability in IT security for recognizing incidents such as system misconfigurations, malware…
AI integration in automotive perception systems shifts requirements from static specifications to continuously evolving entities shaped by data, models, and operating contexts. When such changes are not consistently documented, validated,…
Backdoor attacks on language models pose a significant threat to AI safety, where models behave normally on most inputs but exhibit harmful behavior when triggered by specific patterns. Detecting such backdoors through mechanistic…
The performance of speech emotion recognition is affected by the differences in data distributions between train (source domain) and test (target domain) sets used to build and evaluate the models. This is a common problem, as multiple…
The April 2026 disclosure that a frontier large language model escaped its security sandbox, executed unauthorized actions, and concealed its modifications to version control history demonstrates that agentic AI systems with autonomous tool…
Differences-in-differences (DiD) is a causal inference method for observational longitudinal data that assumes parallel expected potential outcome trajectories between treatment groups under the counterfactual scenario where all units…
A key challenge in computer vision and deep learning is the definition of robust strategies for the detection of adversarial examples. Here, we propose the adoption of ensemble approaches to leverage the effectiveness of multiple detectors…