This paper proposes a method for auditing parliamentary elections to certify a winning majority instead of individual seats, reducing the number of ballots inspected by almost a thousand-fold.

The paper introduces two novel risk-limiting audit techniques—statistical manifest generation and direct ballot selection—that significantly reduce the computational complexity and time required for p…

This paper improves an algorithm for computing lower bounds on the margin of a Single Transferable Vote (STV) election, making risk-limiting audits more practical.

This paper analyzes differential privacy auditing as a bilevel game, showing that naive audit designs fail to detect true harm when developers strategically respond, and proposes an optimal, single-le…

This paper proposes a comprehensive, risk-based auditing framework designed to help internal and external auditors assess the cybersecurity risks posed by diverse IoT devices within corporate and indu…

The paper introduces an optimal black-box auditing framework using Donsker-Varadhan estimators to estimate Rényi differential privacy (RDP) guarantees for machine learning algorithms.

The paper demonstrates that current safety audit metrics are susceptible to strategic platform manipulation, proposing a more robust 'semantic-envelope' metric that better certifies genuine harm reduc…

This paper develops a data-poisoning audit for augmented inverse-probability-weighted estimation to prevent strategic record selection in observational causal analyses.

The paper introduces SafeAudit, a meta-audit framework that systematically enumerates test cases and uses a quantitative metric to uncover significant residual unsafe behaviors in LLM agents that exis…

The paper conducts an interpretability-driven safety audit of eight state-of-the-art LLMs, demonstrating that while interpretability-based steering is a powerful auditing tool, model robustness varies…

This paper characterizes the label complexity of certifying small missed mass in an empirical pipeline and shows that auditing the excluded pool is minimax rate-optimal.

This paper proposes a threshold-free checkpoint audit method using two internal signals to detect if an open-weight checkpoint's refusal mechanism has been stripped.

The paper introduces and demonstrates 'narrow secret loyalties,' a novel type of covert model manipulation that biases model output toward a specific principal's interests under narrow conditions, whi…

The paper introduces the Sequential Triply Robust (STR) estimator, a method that corrects for multiple systematic biases (authorization, reporting, delay, and corruption) in chargeback labels to achie…

The paper introduces an efficient, lightweight LLM framework for smart contract auditing that decouples the audit process into multiple components, achieving high accuracy while significantly reducing…

The paper introduces a comprehensive taxonomy and auditing framework to assess the collective coverage of existing LLM attack benchmarks, revealing significant and systematic gaps in current testing m…

This paper maps the emerging insurability frontier of AI risk by coding 55 AI threat classes against 26 insurance products, identifying four tiers of coverage: affirmative, silent, excluded, and outsi…

The paper introduces lazy Streett supermartingales and their lexicographic extension to certify almost-sure satisfaction of omega-regular properties with polynomial templates under a broad class of sa…

The paper proposes a novel nine-dimension risk assessment framework for institutional DeFi adoption, significantly enhancing existing methodologies by incorporating novel dimensions like composability…

The paper demonstrates that the current per-token billing model for LLMs is susceptible to systematic overcharging because auditing frameworks must rely on evidence provided by the very companies that…