arXiv AI
Jul 20
Coercion and Deception in AI-to-AI Management: An Agentic Benchmark of Unprompted Escalation
arXiv:2607. 15434v1 Announce Type: cross Abstract: Multi-agent systems routinely place one AI agent in authority over another.
What Lies Beyond Perception: A Partition Theory of the Limits of AI Evaluation and Oversight
The paper introduces a partition theory of AI evaluation, arguing that an evaluator’s ability to certify a model’s capabilities is limited by its perception. It shows that if a model’s input partition refines the evaluator’s, it can pass all tests the evaluator can grade, but only a stricter evaluator can raise the discrimination floor. The authors define an echo test that is complete for the evaluator’s tests, proving that passing it only guarantees a minimum discrimination level and cannot detect exact imitation of weaker models.
Related stories
arXiv AI
Aug 20
Competence, Not Accuracy: A Diagnostic for Reference-Free Judge Gates in Skill Optimization
The paper introduces a diagnostic for reference‑free judge gates in text‑space skill optimization. It formalizes a judge as a latent solver, deriving a closed‑form bound on discriminability (ROC‑AUC) in terms of judge competence and answer‑space size, and shows that discriminability is confounded by item difficulty unless a within‑question estimator is used. A non‑intervening probe demonstrates that discriminability is at chance near the competence floor, rises above it, and that the diagnostic can predict gating errors in closed‑loop experiments.
arXiv AI
Sep 3
LLM-as-a-Judge Is Not an Oracle: Why Self-Improving Agents Need Deterministic Guardrails
The paper argues that using a large language model (LLM) as the sole judge in self‑improving agent pipelines is problematic, as the judge can be biased or manipulated, leading to false confidence in system performance. The authors propose a new framework, PROCTOR, which replaces the oracle judge with a deterministic, teacher‑student loop that enforces guardrails such as sandboxing, role separation, and acceptance checks to prevent cheating and ensure reliable evaluation. Experiments across contract analysis, compliance review, and code quality demonstrate that PROCTOR mitigates eleven identified failure modes that previously allowed agents to achieve perfect scores while hiding significant capability gaps.
arXiv Machine Learning
Oct 8
Restricting the Model, Missing the System: Measurement and Accountability in Offensive AI Governance
arXiv:2605.09504v3 Announce Type: replace-cross Abstract: We show that the instruments used to measure AI offensive capability fail in two ways: (i) they overstate harm, and (ii) they credit the mode...
arXiv AI
Jul 31
Adversarial Pragmatics for AI Safety Evaluation: A Diagnostic Framework and Seed Benchmark for Language-Mediated Control
arXiv:2607. 01153v3 Announce Type: replace-cross Abstract: Safety evaluations for language models increasingly depend on judgments about ambiguous natural-language behaviour: whether a model followed an instruction, refused appropriately, complied with a policy, or misreported progress in an agentic task.
arXiv AI
Jun 9
Oversight Has a Capacity: Calibrating Agent Guards to a Subjective, Fatiguing Human
arXiv:2606. 08919v1 Announce Type: new Abstract: As LLM agents begin to take real, irreversible actions (shell commands, file edits, deploys), the standard safety pattern is a human-in-the-loop approval gate: risky actions pause and wait for a person.
arXiv Computer Vision
Oct 2
A Low Grounding Score Is Not an Ungrounded Judge: Identifying the Perceptibility Confound in Multimodal Oversight
arXiv:2610.00111v1 Announce Type: new Abstract: Model judges now supervise multimodal systems at scale, filtering training data, selecting outputs, and supplying the reward that shapes multimodal rea...
arXiv AI
Jun 16
Constitutional Value Potentials: reading and steering internal priority margins in language models
arXiv:2606. 15420v1 Announce Type: cross Abstract: A constitution tells a language model what to value, but little tells us whether it does.
arXiv AI
Oct 2
Verify Claims, Not Scores: Evidence-Based Verification of Modular Agents
The paper proposes a claim‑specific verification audit for modular agents that replaces aggregate task scores with evidence‑based evaluations. Each agent conclusion is recorded with supporting evidence and classified as supported, unsupported, unresolved, or not evaluated, along with the boundary of validity. The audit employs three tools—oracle policies, perfect component replacements, and verifier‑score tests—to trace value changes, locate lost value, and assess verifier effectiveness, demonstrated on a portfolio‑allocation agent in a synthetic market.
arXiv AI
Aug 26
Beyond Accuracy: A Dual-Judge Evaluation Protocol for Vision-Language Models in Legally Grounded Tasks
The paper introduces a dual‑judge evaluation protocol for vision‑language models in legally grounded tasks, pairing a 0‑10 quality judge with a strict binary semantic‑equivalence judge. Using a controlled UK traffic‑sign interpretation task, the authors analyze 4,680 evaluations across visibility and occlusion conditions, finding moderate association between judges and an asymmetric Type II error pattern that is most pronounced under heavy occlusion. The protocol requires only one additional LLM call and reveals quality‑trustworthiness signals that single‑judge methods miss.
arXiv Computation and Language
Oct 22
Conflicted Records: Typed Adversarial Contracts for Auditable LLM Output Quality Enforcement
The paper introduces BiasGuard‑D, a system that uses two adversarial language‑model agents from different families to exchange typed claim documents, with a third model quorum voting to adjudicate. A combined referential‑plus‑semantic invariant validates the verdict, and the iterative deliberation layer shifts remediation from stereotypes to hallucinations, improving right‑answer recovery by 49 percentage points on HaluEval QA. All claims, votes, and instructions are stored as typed records, providing an auditable provenance trail for operators under emerging AI regulations.
arXiv AI
Sep 11
Builder, Defender, Breaker: Measurable Independence and Bounded Autonomy When Generative Models Build, Defend and Test Software
The article discusses how generative models increasingly act as builders, defenders, and breakers of software, challenging the assumption that full autonomy is the ultimate goal. It introduces a framework that defines measurable independence between lifecycle roles based on shared generative substrates, and proposes five autonomy levels, three human roles, and five decision criteria to guide oversight. The authors argue that human authority should focus on specification, accountability, and emergency intervention, and they outline testable hypotheses and protocols to evaluate independence and oversight effectiveness.