This paper identifies a hidden cost of LLM safety alignment: when models are prompted with protected attributes such as race, they often refuse to judge the associated information at all. The authors formalize this as invisibilization using a two-urn Bayesian trust game with an exact benchmark, randomizing only the reporter's description across 6,600+ sessions. They show that some model configurations discard identical evidence far more often for Black-described reporters (46%) than white-described ones (9%), while ironically trusting the assessed Black source slightly more. The result is a probability bias that standard fairness audits miss entirely because they look only at cases where the model actually renders a judgment.

Regulated institutions must produce technical evidence for LLM deployments under the EU AI Act, yet existing benchmark suites are engineer-facing command-line tools while accountable reviewers are non-technical managers. VERA is an open-source, self-hosted dashboard that closes this workflow gap by letting reviewers select a model, run public benchmarks, and read requirement-level scores—with decomposable evidence and run-level energy consumption—in a browser. Its core contribution is wrapping raw measurement into a review and filing process designed for institutional approval.

The paper tests whether open LLMs that a hospital could self-host are ready for clinical use, evaluating four models (8–27B parameters) on the Spanish MIR residency exam under its real scoring and ranking rules. The authors go far beyond headline accuracy to probe calibration, the value of abstention under a wrong-answer penalty, the soundness of clinical reasoning, and whether vision models actually use the images they receive. Their headline message is that a model can rank high enough to choose almost any specialty while remaining overconfident, correct for the wrong reasons, and blind to the clinical plates it is shown.

As LLMs move from isolated tools to networked populations, the structure of who interacts with whom becomes a key safety variable. This paper studies how network topology shapes convention formation and adversarial tipping in populations of up to $10^5$ LLM-derived agents playing the naming game. By placing policies extracted from four models on fully connected, heterogeneous, and sparse regular networks, the authors show that topology modulates convergence speed, can rescue stuck populations, and creates hub-shaped vulnerabilities that well-mixed estimates miss.

This paper introduces a ground-level dataset of 121,536 ground–reasons pairs extracted from 86,464 French civil cassation rulings spanning 2016–2025. Using a deterministic, rule-based pipeline, the authors pair each legal complaint with the Court’s response and classify the pair into established doctrinal families, producing a descriptive grammar of civil cassation. The resource enables quantitative analysis of which grounds succeed, which legal texts are cited, and how both evolve over time, alongside an interactive consultation tool.

Mahran asks what AI evaluation can certify when both examinee and evaluator have perceptual limits. The paper models each agent as a partition over stimuli—inputs it cannot distinguish—and proves a one-sided bound: a sharper examinee can always emulate a coarser evaluator's expected behavior and hide extra capability, while a coarser examinee cannot hide its blind spots under exhaustive probing. This yields predictions about sandbagging, monitor evasion, and weight noise, plus a worrying corollary: exact imitation of a weaker model is undetectable by any behavioural evaluator. The results suggest that behavioural evaluation alone can only establish lower bounds on discrimination, never upper bounds on capability.

Most studies measure how much AI can speed up a task—exposure—rather than whether a worker can actually hand it off. This paper introduces the Agent Delegability Index (ADI), which rates all 18,796 O*NET tasks on a frozen, preregistered four-level delegation ladder, then maps them to ISCO-08 occupations and weights them by employment across 146 countries. By distinguishing partial from full delegation and tagging physical, accountability, trust, and access barriers, the authors reveal a steep income gradient in who can delegate: high-income countries have roughly twice the delegable employment share of low-income ones, driven mainly by the composition of work.

Mental-health chatbots powered by LLMs risk unsafe outputs, yet current safety layers are opaque and developer-driven. This paper proposes a clinician-led guardrail in which two doctoral trainees in clinical psychology author four safety checks, which are then formalized over sixteen typed facts and verified for internal consistency with an SMT solver before deterministic execution. The core claim is not superiority in accuracy but in inspectability: every verdict recomputes from named rules and facts, making disagreements localizable and revisable.

Traditional student models only use responses to standalone questions, ignoring the rich, fine-grained evidence embedded in tutoring dialogues. This paper proposes URF-QD, a framework that rewrites tutor-posed mathematical subtasks as self-contained questions and models them jointly with original items on a shared item response theory scale. The key novelty is a relative-order regularization that uses tutor pedagogical moves to constrain consecutive subtask difficulties, enabling dynamic tracking of student ability across both questions and dialogues. The authors show consistent improvements on two real-world math datasets, though the approach leans heavily on LLM annotations and modest sample sizes.

ReproBAIT evaluates whether sandboxed coding agents can reconstruct published biological AI tools from limited text descriptions when supplied with training data and a fixed compute budget. The authors task frontier LLMs with replicating three tools—a closed-source molecular-design model, ProteinMPNN, and RFdiffusion3—and score functional parity on held-out data rather than source-code similarity. Near-parity is reached on the two simpler tools for only a few dollars, while the most complex tool defeats all agents within the allotted budget, raising questions about memorisation and whether withholding code is enough to control access.

Simulated students let developers test tutoring systems before real learners are available, but LLM role-play suffers from competence leakage because the model substitutes its own knowledge for the learner's. HACHIMI-II counters this by recasting simulation as third-person prediction: it builds a student model, infers a fine-grained learner state for each step, and compiles that state into a deterministic behavior harness that constrains generation. On five datasets, this cuts the ability-distribution gap by 2.4–8.4× and more than doubles error-structure fidelity over the strongest baseline.

This paper confronts the design problem of building dashboards for novel computational techniques whose real-world meaning is still contested. It introduces TopoLA, a system applying Zigzag Persistent Homology to learning-management data, and derives three principles for managing interpretive uncertainty: separating measurement from interpretation, graduated disclosure, and explicit acknowledgment of methodological limits. The argument is timely because fields from explainable AI to bioinformatics increasingly face the same gap between analytical power and interpretive consensus.

Asynchronous online learning structurally fragments learner communities, and cohort-level topological indicators like $\beta_0$ can misleadingly suggest convergence when learners simply drop out. This paper tests whether $\beta_0$ from Zigzag Persistent Homology is participation-sensitive and whether deadline-driven fragmentation–convergence cycles replicate across all 22 OULAD courses. Clarifying this matters because AI-augmented dashboards risk spreading systematically misleading cohesion narratives if they misread attrition-driven $\beta_0$ declines.

Federated learning enables privacy-preserving medical AI but struggles with unreliable uncertainty estimates when clinical data is heterogeneous and imbalanced across sites. TrustFed addresses this by introducing representation-aware conformal prediction, which assigns test samples to calibration clients based on feature-space similarity and aggregates local thresholds via a soft-nearest strategy to provide finite-sample coverage guarantees without centralizing raw data. Validated on over 430,000 images across six distinct imaging modalities, the work advances federated learning from privacy-preserving training toward clinically trustworthy deployment with statistically calibrated uncertainty.

This paper analyzes temporal dynamics in Swiss digital news across French, German, and Italian language regions using a triangulated methodology that combines quantitative NLP with qualitative interpretation. The authors process 1.7 million articles to study how different event types—Brexit, Swiss Wolf, Christmas, and the British Royal Family—are covered across linguistic boundaries, introducing domestication profiles and proximity salience ratios to quantify cultural proximity effects.

This paper introduces the Distributed Human Data Engine (DHDE), a socio-technical framework tackling 'under-vibrancy'—a condition of low visitor density suppressing economic activity—in declining regions like Fukui, Japan. Contrasting with overtourism literature, it integrates Google Business Profile search intent, Japan Meteorological Agency micro-climate data, edge-AI cameras, and 97,719 survey responses to forecast tourism flows and quantify economic leakage. The work promises algorithmic governance via 'dual-nudge' interventions to redirect visitors and coordinate merchant behavior, backed by claims of $R^2=0.810$ explanatory power.

This paper investigates how interrogative stances function as markers of voice and power in French-language digital news. Analyzing over 1.2 million articles from 24 outlets (2023–2024) through a mixed-methods pipeline combining LLM pseudo-labeling and qualitative annotation, the authors operationalize pragmatic concepts like answerhood and dialogicity at scale. The study reveals that questions are sparse but structurally significant, predominantly serving framing functions rather than information-seeking, and centering elite actors over diffuse publics.

This paper tackles the problem of measuring dialectal bias in LLMs for Bengali, a low-resource language with nine major regional variants. The authors propose a two-phase framework combining RAG-based translation to create dialectal benchmarks with an RLAIF-inspired evaluation protocol that uses CoT-first reasoning and multi-judge validation. They expose the catastrophic failure of traditional metrics like BLEU and WER for agglutinative dialectal Bengali, showing that LLM-as-judge better predicts human quality assessments.

WARBENCH is a benchmark for evaluating LLMs in military decision-making, addressing critical gaps in current frameworks by testing International Humanitarian Law (IHL) compliance, edge deployment constraints, fog-of-war robustness, and explicit reasoning. Using 136 high-fidelity scenarios derived from real post-WWII conflicts, the authors expose severe structural flaws: state-of-the-art models collapse under complex terrain and asymmetric force distributions, while edge-optimized models exhibit legal violation rates approaching 70%.