Related papers: Stress-Testing Alignment Audits With Prompt-Level …
The rapid advancement of Vision-Language Models (VLMs) has brought their safety vulnerabilities into sharp focus. However, existing red teaming methods are fundamentally constrained by an inherent linear exploration paradigm, confining them…
Adversarial decision-making in partially observable multi-agent systems requires sophisticated strategies for both deception and counter-deception. This paper presents a sequential hypothesis testing (SHT)-driven framework that captures the…
LLMs are increasingly embedded in everyday decision-making, yet their outputs can encode subtle, unintended behaviours that shape user beliefs and actions. We refer to these covert, goal-directed behaviours as hidden intentions, which may…
Existing red-teaming studies on GUI agents have important limitations. Adversarial perturbations typically require white-box access, which is unavailable for commercial systems, while prompt injection is increasingly mitigated by stronger…
Structured data-quality issues, such as missing values correlated with demographics, culturally biased labels, or systemic selection biases, routinely degrade the reliability of machine-learning pipelines. Regulators now increasingly demand…
Humans are black boxes -- we cannot observe their neural processes, yet society functions by evaluating verifiable arguments. AI explainability should follow this principle: stakeholders need verifiable reasoning chains, not mechanistic…
Search-based testing is widely used to find bugs in models of complex Cyber-Physical Systems. Latest research efforts have improved this approach by casting it as a falsification procedure of formally specified temporal properties,…
Science and technology have a growing need for effective mechanisms that ensure reliable, controlled performance from black-box machine learning algorithms. These performance guarantees should ideally hold conditionally on the input-that is…
Standard safety alignment optimizes Large Language Models (LLMs) for universal helpfulness and honesty, effectively instilling a rigid "Boy Scout" morality. While robust for general-purpose assistants, this one-size-fits-all ethical…
LLM-based agents execute real-world workflows via tools and memory. These affordances enable ill-intended adversaries to also use these agents to carry out complex misuse scenarios. Existing agent misuse benchmarks largely test…
Large language model (LLM) based agents are increasingly used to automate financial transactions, yet their reliance on contextual reasoning exposes payment systems to prompt-driven manipulation. The Agent Payments Protocol (AP2) aims to…
We consider the issue of strategic behaviour in various peer-assessment tasks, including peer grading of exams or homeworks and peer review in hiring or promotions. When a peer-assessment task is competitive (e.g., when students are graded…
Large language model (LLM) agents increasingly rely on skills to package reusable capabilities through instructions, tools, and resources. High-quality skills embed expert knowledge, curated workflows, and execution constraints into agents,…
System prompt configuration can make the difference between near-total phishing blindness and near-perfect detection in LLM email agents. We present PhishNChips, a study of 11 models under 10 prompt strategies, showing that prompt-model…
Manufacturing quality audits are pivotal for ensuring high product standards in mass production environments. Traditional auditing processes, however, are labor-intensive and reliant on human expertise, posing challenges in maintaining…
Safe deployment of Large Language Model (LLM) agents in autonomous settings requires reliable oversight mechanisms. A central challenge is detecting scheming, where agents covertly pursue misaligned goals. One approach to mitigating such…
We study secret elicitation: discovering knowledge that an AI possesses but does not explicitly verbalize. As a testbed, we train three families of large language models (LLMs) to possess specific knowledge that they apply downstream but…
Performance modeling typically relies on two antithetic methodologies: white box models, which exploit knowledge on system's internals and capture its dynamics using analytical approaches, and black box techniques, which infer relations…
We study the problem of auditing the fairness of a given classifier under partial feedback, where true labels are available only for positively classified individuals, (e.g., loan repayment outcomes are observed only for approved…
This work presents an attack-aware deepfake and image-forensics detector designed for robustness, well-calibrated probabilities, and transparent evidence under realistic deployment conditions. The method combines red-team training with…