Adaptive auditing of AI systems with anytime-valid guarantees

arXiv:2605.11496v1 Announce Type: new Abstract: Recent published evidence from frontier laboratories shows that contemporary AI models can recognise evaluation contexts, latently

arXiv:2608.00794v1 Announce Type: new Abstract: Agentic AI systems are evaluated using automated benchmarks whose scores justify deployment decisions, safety certifications, and

arXiv:2609.10873v1 Announce Type: new Abstract: Independent evaluation can reject harmful policy updates yet also prevent useful continual learning. We argue that update admission

arXiv:2610.21581v1 Announce Type: new Abstract: Mathematical agents are commonly evaluated by answer accuracy or proof acceptance. We introduce ClaimScope, a 72-task benchmark for

arXiv:2606.23927v1 Announce Type: new Abstract: Agentic AI systems powered by large language models (LLMs) are rapidly evolving into autonomous decision-making systems, exposing

arXiv:2605.16198v1 Announce Type: new Abstract: We examine one particular dimension of AI governance: how to monitor and audit AI-enabled products and services throughout the AI

arXiv:2608.23817v1 Announce Type: new Abstract: SHAP and LIME are now standard tools for interpreting black-box predictions, yet their outputs can vary substantially when the

arXiv:2604.18789v1 Announce Type: new Abstract: Reinforcement Learning from Human Feedback (RLHF) is central to aligning Large Language Models (LLMs), yet it introduces a critical