Who judges the judges? Governance from metrics: a runtime framework for continuous LLM compliance monitoring

• URL: http://arxiv.org/abs/2605.24737v1
• Date: Sat, 23 May 2026 21:21:33 GMT
• Title: Who judges the judges? Governance from metrics: a runtime framework for continuous LLM compliance monitoring
• Abstract summary: Current approaches to AI compliance treat conformity as a binary, audit-time verdict.<n>We argue that this compliance fiction is structurally ill-suited to the requirements of the EU AI Act.<n>We introduce governance from metrics, a principle whereby regulatory compliance is derived as a continuous signal from behavioural observability.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Current approaches to AI compliance treat conformity as a binary, audit-time verdict rather than a continuous, measurable property of production systems. We argue that this compliance fiction is structurally ill-suited to the requirements of the EU AI Act, which demands ongoing human oversight and the detection of emergent behavioural drift in deployed systems. We introduce governance from metrics, a principle whereby regulatory compliance is derived as a continuous signal from runtime observability rather than from static assessments. Building on this principle, we present govllm, an open-source framework implementing a governance-driven routing architecture in which model selection is determined by accumulated compliance scores rather than by latency or cost alone. Central to our approach is a panel of regulatory judges - LLM evaluators specialised per criterion (EU AI Act, GDPR, ANSSI, accessibility) - whose inter-judge disagreement we reframe not as noise but as a regulatory uncertainty signal warranting human arbitration. We validate this approach through a ground truth corpus of 49 annotated prompt/response pairs across five regulatory criteria, evaluated by four small language models (SLMs, 1.7B-7B parameters) running fully on-premise. Agreement rates range from 51.5% (mistral:7b) to 69.1% (phi4-mini), with no single model dominating across all criteria - empirically motivating the Profile-as-jury design. We further document three structural failure modes in small regulatory judges and a judge-specific position bias that degrades agreement by up to 25 percentage points across three question-order conditions (original, reversed, permuted). govllm is released as open-source software to support reproducible AI governance research.

Related papers

• AspisAI: A Canonical, Machine-Interpretable Governance Framework for Automated Multi-Standard Compliance Monitoring [0.0]
  This paper presents emphAspisAI, a bounded, standard-agnostic governance framework that translates selected requirements into a canonical, machine-interpretable control model.<n>Within a bounded scope of 26 representative requirements, the framework is evaluated in a controlled simulation against five governance-oriented criteria.<n>Results include full requirement encoding, 88.5,% mapping coverage, complete traceability, and correct detection of all introduced gaps.
  arXiv  Detail & Related papers  (2026-09-09T22:37:12Z)
• Grounded Normative Rule Generation with Structured Search [47.28595196084396]
  We formalize the problem as Grounded Normative Rule Synthesis (GNRS)<n>We introduce GNRS-Search, a framework that utilizes Markov Chain Monte Carlo (MCMC) sampling to optimize a discrete, five-slot And-Or Graph (AOG)<n>We evaluate our approach on GNRS-Bench, a benchmark spanning 116 controlled goals across eight scene families, and RealCharterBench, which evaluates transfer to 53 real-derived policy tasks with hidden source clauses.
  arXiv  Detail & Related papers  (2026-08-23T05:43:42Z)
• Time to REFLECT: Can We Trust LLM Judges for Evidence-based Research Agents? [61.49434544687523]
  We introduce REFLECT, a meta-evaluation benchmark targeting fine-grained failure detection in agentic environments.<n> REFLECT defines a detailed taxonomy of process- and outcome-level failure modes, instantiated by performing controlled and localized interventions.<n>Our experiments show that even the best-performing models achieve overall accuracies below 55% across reasoning, tool-use, and report-quality failures.
  arXiv  Detail & Related papers  (2026-05-18T23:55:08Z)
• Designing escalation criteria for international AI incident response: criteria, triggers, and thresholds [0.0]
  No operational criteria exist for determining when a detected AI incident warrants escalation beyond national handling to international coordination.<n>This paper proposes an escalation framework to address this gap, intended as a common reference point across jurisdictions.
  arXiv  Detail & Related papers  (2026-04-25T07:31:58Z)
• Governed Reasoning for Institutional AI [0.0]
  Cognitive Core is a governed decision substrate built from nine typed cognitive primitives.<n>Human review is a condition of execution rather than a post-hoc check.<n>We benchmark three systems on an 11-case balanced prior authorization appeal evaluation set.
  arXiv  Detail & Related papers  (2026-04-12T14:09:18Z)
• Automated Self-Testing as a Quality Gate: Evidence-Driven Release Management for LLM Applications [51.56484100374058]
  We present an automated self-testing framework that introduces quality gates with evidence-based release decisions.<n>We evaluate the framework through a longitudinal case study of an internally deployed multi-agent conversational AI system.
  arXiv  Detail & Related papers  (2026-03-13T20:44:15Z)
• Mirror: A Multi-Agent System for AI-Assisted Ethics Review [104.3684024153469]
  Mirror is an agentic framework for AI-assisted ethical review.<n>It integrates ethical reasoning, structured rule interpretation, and multi-agent deliberation within a unified architecture.
  arXiv  Detail & Related papers  (2026-02-09T03:38:55Z)
• Compliance as a Trust Metric [1.0264137858888513]
  This paper bridges this research gap by operationalizing regulatory compliance as a quantitative and dynamic trust metric.<n>Our contribution is a quantitative model that assesses the severity of each violation along multiple dimensions, including its Volume, Duration, Breadth, and Criticality.<n>We evaluate ACE on a synthetic hospital dataset, demonstrating its ability to accurately detect a range of complex HIPAA and HIPAA violations.
  arXiv  Detail & Related papers  (2026-01-03T21:14:40Z)
• Judging by the Rules: Compliance-Aligned Framework for Modern Slavery Statement Monitoring [24.13989765643719]
  Modern slavery affects millions of people worldwide, and regulatory frameworks such as Modern Slavery Acts now require companies to publish detailed disclosures.<n>These statements are often vague and inconsistent, making manual review time-consuming and difficult to scale.<n>We propose a novel framework that harnesses AI for rule-level compliance verification while preserving expert oversight.
  arXiv  Detail & Related papers  (2025-11-11T03:41:44Z)
• Making LLMs Reliable When It Matters Most: A Five-Layer Architecture for High-Stakes Decisions [51.56484100374058]
  Current large language models (LLMs) excel in verifiable domains where outputs can be checked before action but prove less reliable for high-stakes strategic decisions with uncertain outcomes.<n>This gap, driven by mutually cognitive biases in both humans and artificial intelligence (AI) systems, threatens the defensibility of valuations and sustainability of investments in the sector.<n>This report describes a framework emerging from systematic qualitative assessment across 7 frontier-grade LLMs and 3 market-facing venture vignettes under time pressure.
  arXiv  Detail & Related papers  (2025-11-10T22:24:21Z)
• Zero-shot reasoning for simulating scholarly peer-review [0.0]
  We investigate a deterministic simulation framework that provides the first stable, evidence-based standard for evaluating AI-generated peer review reports.<n>First, the system is able to simulate calibrated editorial judgment, with 'Revise' decisions consistently forming the majority outcome.<n>Second, it maintains unwavering procedural integrity, enforcing a stable 29% evidence-anchoring compliance rate.
  arXiv  Detail & Related papers  (2025-10-02T13:59:14Z)
• J4R: Learning to Judge with Equivalent Initial State Group Relative Policy Optimization [69.23273504123941]
  We train judges to be robust to positional biases that arise in more complex evaluation settings.<n>We introduce ReasoningJudgeBench, a benchmark that evaluates judges in diverse reasoning settings not covered by prior work.<n>We train Judge for Reasoning (J4R), a 7B judge trained with EIS-GRPO that outperforms GPT-4o and the next best small judge by 6.7% and 9%.
  arXiv  Detail & Related papers  (2025-05-19T16:50:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.