The paper introduces Policy-as-Skill (PaS), a modular runtime that packages evidence validation, review routing, version control, and auditability as executable, versioned policy capabilities for large language models. PaS was evaluated on 600 development tasks using a Gemma4 backend, achieving 53.8% exact accuracy and high audit completeness, and deterministic control further improves aggregate accuracy to 61.2% in a task‑dependent manner.
arXiv:2609.37457v1 Announce Type: new Abstract: Enterprise artificial-intelligence agents increasingly call tools, modify infrastructure, and process protected data, creating a need to separate actio...
The paper introduces Runtime Assurance Contracts (RAC) as a formal policy framework for high‑risk AI agents, addressing the "assurance‑transition gap" by binding autonomy boundaries, component eligibility, evidence state, transition policy, human‑review capacity, and non‑compensatory gates. RAC allows soft metrics to influence routing while mandating retries, switches, escalations, deferrals, or stops when mandatory gates fail or are unknown, ensuring aggregate performance cannot alone authorize action. The authors define the contract, evidence record, permission rule, and five invariants, and evaluate RAC through deterministic failure‑injection studies, hand‑authored traces, and a prospective synthetic holdout, comparing it to score‑only and restricted protocol baselines.
arXiv:2606. 14581v1 Announce Type: cross Abstract: Granting LLMs direct control over costly, irreversible scientific experiments leads to unsafe exploration and unstable performance, but discarding LLM creativity entirely sacrifices significant optimization potential.
FinRiskAtlas is a Chinese-language benchmark designed to evaluate large language models (LLMs) for financial risk review by focusing on decision‑aligned tasks rather than generic financial knowledge. It contains 9,742 instances across 53 task families, including 42 domain‑knowledge families and 11 downstream review operations defined by explicit evaluation contracts. The extended FinRisk‑Ask framework replays 680 pre‑action states from 104 professional trajectories, withholding future evidence during inference to assess evidence‑state control and request targeting. Results across 33 model configurations show that operation‑level evaluation yields distinct rankings and that knowledge‑based shortlisting can incur significant regret, while frequent use of the Ask branch does not necessarily improve evidence acquisition, highlighting gaps in broad financial capability scores.
The paper introduces Governance-as-Code (GaC), a framework that translates the EU AI Act’s technical requirements into 43 machine‑checkable acceptance criteria across six compliance modules. GaC runs within a CI/CD pipeline, producing Article‑indexed audit evidence and providing actual Rego policy code. The authors validate GaC on two enterprise deployments, showing it reproduces manual audit findings—including three penalty‑triggering violations—while reducing audit labor by about 75%.
arXiv:2607. 09682v1 Announce Type: new Abstract: AI systems are increasingly used to assist consequential decisions in regulated domains such as auditing, finance, and healthcare.
The paper investigates how procedural traces—detailed step-by-step accounts of a language model’s reasoning—affect the decision-making of LLM overseers tasked with auditing another model’s outputs. Using signal detection theory, the authors evaluated five overseers on 19 compliance tasks, finding that while error detection remains high when disconfirming evidence is always visible, more elaborate traces shift the decision criterion toward rejection, leading to increased false alarms. The study also shows that providing option labels reduces the stated inability to link evidence to options, yet some overseers still exhibit residual rejection of correct work that grows with trace detail.
The paper investigates how personalized agents decide to use, ignore, update, or query retrieved user memory before acting on a task. An empirical audit protocol is developed to test structured intermediate outputs, revealing that while exposing state definitions improves accuracy, an explicit state-output field does not significantly enhance policy accuracy for large language models. The study also shows that example-level accuracy overstates consistency, with full four‑way family success being rare, and that providing benchmark‑associated state labels merely conditions predictions rather than proving internal fidelity.
The paper introduces the twin‑prefix framework to evaluate how the size of the verification unit—i.e., how many actions a pre‑execution LLM monitor reviews in one call—affects its performance. By pairing each gold plan with a twin that differs by a single write and injecting a controlled error, the authors isolate the impact of review length on catch rates and false rejections. Their findings show that longer review windows increase rejection rates but do not improve discrimination, with the highest informedness occurring at one or two actions across all judges and domains.
The paper introduces the Alignment Flywheel, a governance‑centric hybrid multi‑agent system (MAS) that separates decision generation from safety governance. It defines a Proposer that generates candidate trajectories, a Safety Oracle stack that evaluates safety, and an Enforcement layer that applies risk policies at runtime. A governance MAS oversees monitoring, red‑teaming, verification, and versioned release management, enabling patch‑local fixes to safety failures without retraining the Proposer. The architecture is implementation‑agnostic and is demonstrated in two scenarios: a learned spatial Oracle and a clinical GenAI proxy. The authors provide open‑source code at https://github.com/decide-ugent/Alignment-Flywheel.
arXiv:2608. 16852v1 Announce Type: new Abstract: Regulatory compliance monitoring in deployed language models is increasingly implemented as a legal and audit control, checking model outputs against written rules spanning data protection, healthcare, financial regulation, and platform policy.
arXiv:2609.14400v1 Announce Type: new Abstract: Agent benchmarks evaluate policy compliance but assume each policy determines a unique correct action. Natural-language policies can violate this assum...