Papers

We present a robust benchmark for measuring cultural value orientation in large language models on the two Inglehart-Welzel axes over several domains and roles (hence GYROval - Gridded Yielding of Robust value Orientation), together with the results of administering it to twenty models. Items are binary contrastive scenarios in the sense introduced by CDEval: both options are legitimate courses of action, neither is correct, there is no answer key, and a model's score on an axis is the proportion of its responses falling on the counted pole. Eleven of the twenty models were additionally administered a paired Russian translation of the identical items and a second sampling temperature. The instrument is publicly released in both languages. Stability was assessed by treating the vignette as the unit of analysis, ranking the models within the levels of each perturbation factor, and summarising the agreement between levels by tie-corrected Kendall's W against an empirical permutation null.

When someone asks an AI assistant which doctor to see or which firm to trust with their savings, the answer is a referral. We audit AI provider recommendations in four registry-backed service domains across the 100 largest U.S. metropolitan areas, matching every recommendation against the official registry for its domain (Medicare clinician and facility records, and SEC adviser disclosures), under three conditions: an open-weight model, a proprietary model without web search, and the same proprietary model with search. Without search, both models largely fabricate recommendations in the domains the web covers thinly. Only 4% of the open-weight model's recommended doctors and 11% of the proprietary model's match a clinician in the queried city, and the open-weight matches are name coincidences: its matched clinicians are no likelier to be primary-care doctors than names drawn at random from the registry. With search, 64-71% of recommendations in the same domains match a real provider. Search also changes who is recommended. Without it, recommended advisory firms carry SEC misconduct disclosures at 3.6 times the registry base rate, even after adjusting for firm size; with search, significantly below it. Restaurants, where quality and visibility are separately measurable, show a 3-5x review-count premium but a rating premium of at most a tenth of a star. Finally, search largely removes the metro-size penalty: without it, real recommendations concentrate in the largest metros; with it, match rates are similar across metro-size terciles. Whether an AI referral is trustworthy depends strongly on its retrieval configuration rather than on the underlying model alone, yet an answer produced without retrieval often carries no sign that its recommendations were never verified.

Organizations increasingly use oversight loops where one large language model (LLM) audits another's outputs alongside procedural traces of claimed steps. A common concern about such LLM-as-a-judge pipelines is that detailed traces make overseers gullible. Using signal detection theory, we audit five LLM overseers on 19 compliance tasks (4,551 analyzed judgments), varying only trace detail and evidence labeling. With disconfirming evidence always visible, error detection remains near ceiling. Instead, elaborate traces shift the decision criterion toward rejection, increasing false alarms in susceptible overseers. Without option labels, human-validated reason coding shows about 60% of false alarms cite an inability to tie evidence to its option. Labels eliminate this stated reason, yet residual rejection of correct work persists in those overseers and rises with trace detail. Procedural traces thus act as governance artifacts that shape oversight decisions. AI auditors should be evaluated by their decision criterion and false-alarm behavior, alongside accuracy.

In recent years, large language model-based AI companion and role play chatbots have grown increasingly popular. People turn to these chatbots for emotional support and to engage in romantic and erotic role play. Although prior research suggests that digital role play can help people explore their gender and sexuality, LLM based technologies are also replete with gender and sexuality biases. In this study, we examine one way that AI chatbots can harm users: misgendering. In order to study chatbot misgendering we qualitatively analyzed 326 posts mentioning misgendering that were shared in AI companion or role play subreddits. We document how chatbot misgendering takes place and how, in response, users engage in ongoing work to curate their gender presentation to prevent and repair misgendering. We discuss how researchers and designers can mitigate chatbot misgendering and consider the implications of using AI chatbots for identity exploration.

Open-weight large language models are rapidly entering hiring pipelines, yet their discriminatory failure modes – and the regulatory exposure these create under the EU AI Act high-risk classification (Annex III) and U.S. EEOC adverse-impact analysis – remain poorly understood. We present the first systematic, multi-model audit of open-weight LLMs that treats job-posting language as the primary experimental variable, evaluating six models (Llama 3.2, Mistral, Gemma 3, Qwen 3, Phi 3, DeepSeek-R1) across four controlled experiments that jointly probe recruiter-simulation and job-seeker-simulation tasks. We find that (1) agentic posting language depresses recruiter recommendation scores for female candidates (r_rb = 0.309, p_Bonf = 7x10^-5; model-fixed-effects r_rb = 0.448), while communal language partially reverses the penalty; and (2) coded-exclusion language suppresses non-White recruiter scores at large effect sizes (r_rb = 0.646-0.758) and, on the job-seeker side, selectively deters non-White personas from expressing interest – operationalizing a chilling-effect mechanism at scale. A label-ablation experiment isolates the explicit demographic persona label as the primary causal driver, and Word Embedding Association Tests corroborate these findings at the representational level (d = 1.01-1.45 under Caliskan et al.'s multi-word gender attribute lists). We translate these results into a concrete pre-deployment audit protocol – posting-vocabulary scoring, persona-conditioned LLM probing, and adverse-impact flagging against the four-fifths threshold – that operationalizes the documentation and risk-management obligations Annex III imposes on high-risk AI in recruitment.

K-12 robotics and AI education remains difficult to scale, especially in rural regions lacking sustained technical mentorship. Programs like FIRST provide competition pathways and instructional opportunities, but they do not eliminate the need for local programming and robotics expertise. We introduce AI, Robotics, & Community (ARC), a hubs-based framework where colleges train undergraduate mentors and host workshops for nearby K-12 teams. Mature school programs can become secondary hubs that support additional schools, creating a self-reinforcing education loop where mentorship reach propagates geographically and can even grow super-linearly. We first evaluate ARC through a trial deployment at one university. The trial created three rural robotics teams. On five-point Likert surveys, mean increases in K-12 programming knowledge, resource access, and practice opportunities were 2.00, 2.25, and 1.25 points. Likewise, undergraduate confidence teaching technical concepts, adapting explanations, managing groups, and finding mentoring enjoyable and meaningful increased by 1.29, 1.14, 1.00, and 1.14 points. Additionally, we create a spatial Markov model of ARC's growth and simulate it using the state of Indiana as a testbed. Under moderate conditions, we find that ARC reaches 74% of Indiana's 1,925 public K-12 schools and produces 992 robotics programs after 40 years, compared with 161 projected under natural growth alone. Together, these results show ARC can create and support rural robotics programs, train undergraduate AI and robotics mentors, and potentially scale mentorship across a region.

AI is increasingly used to support decarbonization decisions across the built environment, yet the development, training, and use of AI consume energy and induce CO2e emissions. However, existing assessments often report physical-system savings while omitting AI-side emissions. Moreover, they rarely account for the mismatch between when AI costs occur and when decarbonization benefits materialize, which may be substantial for infrastructure-scale projects. To address these issues, we present a time-aware assessment framework that models avoided emissions and AI-induced emissions as discrete-time streams over a finite time horizon. In demonstrating this process, we seek to show that time-aware assessment can support temporal decision-making, identify cases in which accounting for time value of carbon can change preferred rankings relative to time-invariant totals, and explore how decisions may vary with slightly different governance priorities. Using four representative interventions with intentionally different temporal profiles (multi-project low-carbon concrete design support, AI-assisted construction logistics, agentic HVAC control, and predictive maintenance), we demonstrate how discounting can change preferred rankings relative to time-invariant totals and supports ranking sensitivity analysis, discounted payback screening, and break-even discount-rate analysis. We also provide decision guidelines that support go/no-go screening, timing decisions, and minimum "bang-for-your-buck" thresholds. Ultimately, this work contributes a lightweight framework for deciding whether and when to deploy AI-enabled interventions for decarbonization under explicit time preference.

Text-to-image generative models are widely used in professional and creative settings, yet how they represent gender across occupations – and whether newer models are fairer – remains poorly understood across multiple generations. We evaluate gender representation across 20 occupations, 5 prompt templates, and 4 Stable Diffusion model generations (SD 1.5, SD 2.1, SDXL, SD 3 Medium), generating 8,000 images with n = 100 per occupation-model cell (5 prompts x 20 images), and classifying all with DeepFace. Across the 8,000 open-source images, 76.4% show male subjects (95% CI [75.1%, 78.7%], p < 2.2 x 10^-16, Benjamini-Hochberg adjusted). More strikingly, 57.6% of images for historically female-coded occupations show male subjects (raw p = 3.43 x 10^-22, BH-adjusted p = 1.71 x 10^-21). All nine significant tests reported in this paper survive BH correction across 10 tests. When compared against U.S. Bureau of Labor Statistics workforce data, models underrepresent women by 20-46pp on average, with particularly large deviations for near gender-balanced occupations: scientist (48% female in BLS, 82-99% male in model outputs) and cleaner (46% female in BLS, 80-92% male in outputs). Model generations do not improve steadily: bias worsens from SD 1.5 to SDXL before partially recovering in SD 3 Medium. A preliminary comparison with GPT-image-1 on five occupations suggests lower bias than open-source models, though the practical effect is small (Cramer's V = 0.080) and the comparison is exploratory. No model achieves gender parity.

AI is changing what leaders must judge, explain, learn, and coordinate, yet existing measures do not capture these behaviors at the level needed to study leadership in AI-enabled work. We develop the AI Leadership Battery, which organizes 36 behaviorally specific subdimensions into 11 theory-specified content families. Following established scale-development procedures, the research used deductive item generation; content validation of definitional correspondence and definitional distinctiveness; exploratory factor analysis and item reduction; confirmatory factor analysis in independent samples; and tests of internal consistency reliability, convergent validity, discriminant validity, and criterion-related and incremental validity. Across the development and validation studies, the analyses provided evidence for the Battery's content, multidimensional structure, reliability, and distinction from selected orbiting constructs. The Battery also contributed additional information beyond orbiting constructs across organizational growth, decision speed, customer/stakeholder response capability, AI-enabled team performance, AI-enabled work experience, AI security and risk management, and AI adoption and integration. The resulting measure provides researchers with a behavioral framework for examining how leaders in AI-enabled work regulate judgment, learning, adaptation, transparency, and accountability.

Poor indoor air quality can cause up to five times more direct health problems to occupants than outdoor air. In particular, it may cause headaches, fatigue, eye/throat irritation, and long-time exposure is linked to respiratory and heart as well as some forms of cancer. Despite the importance of indoor health and well-being, most current monitoring devices and systems (usually for offices and workspaces) are passive. The Environmental Quality Monitor (EnQyMo) platform is a generic Internet of Things (IoT) middleware designed to process several sensor data related to air quality in indoor spaces and correlate this data with health exposure risks of users/workplace employees. Using Bluetooth Low Energy (BLE) beacons and a mobile IoT middleware it is able to identify the (smartphone) users exposed to these polluted air or high CO2 (carbon dioxide) levels, and generate location-specific alarms only to the users at the places with the unhealthy air conditions. At the core of EnQyMo is an agency of Large Language Models (LLMs) capable of interpreting regulatory standards and scientific literature to automatically identify critical health exposure levels.

An important component of urban accessibility, particularly for wheelchair users and people with reduced mobility, is sidewalk compliance with measurable requirements. We test whether effective width, longitudinal slope, cross slope, and pavement condition can be assessed reliably from pedestrian-level imagery using vision-language models (VLMs). We present the first application of sampling-based conformal prediction (CP) for VLM-based accessibility assessment. We evaluate four VLMs on 514 sidewalk images from Seoul, South Korea, with field-measured ground truth. Conformal calibration attains the nominal 90% coverage for all models and attributes, but the calibrated regions differ in informativeness. Effective width yields the most informative estimates, with a mean interval half-width of about 1.0 m for the best model. Since every model overestimates width, asymmetric calibration shortens the intervals by up to 33% at unchanged coverage. Longitudinal slope is marginally informative, cross-slope intervals are too wide to resolve regulatory thresholds, and pavement-condition sets degenerate to all five grades (A-E) for three of the four models. Uncalibrated intervals from raw sampling dispersion cover only 17-47% of field-measured values at a nominal 90% level. Among the images with the most self-consistent responses, these intervals miss the field-measured value in up to 96% of cases. Response self-consistency is therefore not evidence of accuracy, and sampling dispersion cannot be interpreted as uncertainty until it has been calibrated against field-measured ground truth. No quantitative attribute reaches the precision required for general compliance assessment, but CP identifies from calibration data alone which attributes can support screening of segments far from the thresholds. We release the annotated pedestrian-level images and their corresponding field-measured attribute values.

Who the judge is can affect an LLM-as-judge result, but measuring that effect without confusing it with candidate quality is difficult. We study four open-weight families (Llama 3.1, Qwen 2.5, Gemma 2, and Yi 1.5) in a fully crossed pairwise design with 9,312 judgments. A common per-family statistic is strongly confounded with candidate quality and correlates with Bradley-Terry ability at r = 0.95. We derive a corrected estimator that holds the candidate family fixed and compares judges. All four families then show a positive same-family lift (3.4-8.4 percentage points), with global FPS 0.067 (95% CI [0.053, 0.084], permutation p = 0.0002). The effect remains under panel-based quality controls, an independent human-consensus anchor, and a float16 judging replication. Judge-side likelihood is closely related to the effect: adding likelihood advantage reduces the controlled coefficient by 61%, which we treat as descriptive attenuation rather than causal mediation. Position is a separate failure mode. Across the panel, 55.4% of AB/BA pairs reverse, and reversal above 50% is incompatible with a simple independent content-noise model. Relative to a family-balanced reference, panel composition changes 18.5% of pairwise outcomes. A complete reproducibility archive has been prepared for public release.

AI systems have begun to report solutions, disproofs, and substantive advances on long-standing mathematical problems, raising questions about whether they approach research in the same way as mathematicians. This study compares public AI research accounts with the human literature on 11 such problems. The human corpus contains 58 papers that directly addressed the same mathematical targets later reported by AI sources as resolved, disproved, or substantially advanced; 31 within-problem comparisons were constructed from these materials. Six validated text-based measures capture problem resolution, method articulation, uncertainty and boundary specification, successor-question generation, generality, and cross-disciplinary integration. AI accounts place greater emphasis on resolving the focal problem and connecting ideas across fields. Human papers devote significantly more attention to explaining methods, specifying assumptions and limitations, and identifying questions for subsequent research. No precise difference is detected in generality. The estimated directions remain unchanged when each mathematical problem is removed in turn. The findings reveal two distinct research profiles: AI accounts concentrate on closing and recombining problems, whereas mathematical papers more extensively document the procedures, limits, and research opportunities through which results become cumulative knowledge. Evaluating research AI therefore requires attention to the organization of inquiry, not only whether a target is solved.

Digital trace data provide rich measures of behavior in everyday settings, but the research ecosystem supporting their collection is constrained by declining platform API access and a historical reliance on publicly observable data. Participant-mediated data donation offers a complementary approach in which individuals contribute selected portions of their own digital histories to research. Such data can include longitudinal and non-public behavior, span multiple platforms and modalities, and be linked to independently collected study measures, enabling study designs that are difficult to implement using public social media data alone. These opportunities also introduce methodological challenges around participant control, data minimization, heterogeneous platform exports, privacy, and governance, particularly when semantic or multimodal content is necessary to study the construct of interest. We present CANDOR (Collecting and Analyzing Networked Data for Open Research), an end-to-end infrastructure for participant-mediated collection and governance of sensitive digital trace data. CANDOR supports participant-directed selection of platforms, data types, and temporal ranges; modular platform- and modality-specific parsing and de-identification; linkage to independent study measures; and protected processing, storage, and access. We derive design requirements for this class of research and compare CANDOR with existing data donation infrastructures, identifying how different approaches support participant control, data minimization, scientifically necessary data richness, study-design flexibility, and governance. Together, this work provides a methodological and infrastructural framework for using participant-contributed digital traces in behavioral research, particularly when the data needed to address a scientific question are longitudinal, non-public, multimodal, or sensitive.

Multi-agent LLM systems increasingly automate tasks involving agents with different levels of privilege, creating a security risk in which a compromised low-privilege agent can influence a higher-privilege agent and trigger an unauthorized action. We study attack propagation in a four-agent LangGraph pipeline comprising a Supervisor, Researcher, Validator, and Executor. We evaluate shared-memory poisoning and indirect prompt injection through a forged approval embedded in a retrieved document. We compare the Validator's judgment with an independent authorization layer using task-bound signed tokens and a separately verified policy oracle. Our contribution is an empirical study of attack propagation, a component-level ablation of the authorization boundary, and the Judgment Bypass Rate (JBR), which measures compromise at the attacked agent rather than at the final action. Across three seeds and 60 labeled tasks, memory poisoning reaches execution in every undefended trial. With authorization enabled, it achieves 100% JBR but 0% Unsafe Action Rate, showing that the Validator can remain compromised while execution is contained. Against an attacker possessing the signing secret, the policy oracle provides the observed containment, while an independently authored least privilege policy preserves this result. An Observer layer reduces the false-positive rate for agent hijacking from 49% to 7% without weakening execution-level security. These results show that structural authorization can contain compromised agent behavior even when upstream LLM judgment fails.

Domain-specialized language models are widely used for scientific question answering, but stronger general-purpose systems raise a sharper question: when does domain-specific fine-tuning remain valuable for open-ended scientific reasoning? We study this in astronomy with a curated QA benchmark from publicly available 2017–2026 Olympiad-style materials. The free-response subset contains 300 questions, including 204 text-only and 96 image-linked examples. We compare open-weight and API-served general-purpose, multimodal, and astronomy-specialized models using judge-based correctness and complementary reference metrics. Strong general-purpose models establish the highest correctness baseline in this testbed, while analyses of metric agreement, judge sensitivity, benchmark composition, and modality reveal variation not captured by a single leaderboard. These results motivate treating domain specialization as a task- and deployment-dependent property and highlight the role of domain-specific evaluation in determining which models, capabilities, and evaluation criteria are appropriate for scientific workflows.

AI-assisted claims can appear authoritative when evidence, analysis, human authorization, presentation, and correction history refer to different states. Provenance, attestation, and transparency expose history but alone do not specify the publication transition examined here. We develop Publication Authority as an exact-state, non-transferable, single-use publication capability and instantiate it in PAC-2026 (Publication-Accountability Calculus), a machine-readable AIJIM Protocol candidate. We evaluate its fourth bounded semantic freeze (SF-4), a fixed-profile specification designed for replaceable bindings. Six obligations govern evidence, runs and artifacts, measurement disclosure, authorization, surface correspondence, and lifecycle continuity. Each yields a target-bound witness, localized counterexample, or localized unverifiability; none can compensate for another. Only a fresh, complete all-pass record derives the permit consumed by one atomic publication transition. We use identity vectors, adversarial cases, finite models, and historical implementations. Ten models explored 110,764 safe reachable states; 76 unsafe configurations produced the expected violation or observer countermodel. A reader surface passing its correspondence check cannot authorize publication unless the accepted record admits that surface. SF-4 separates evidence horizon from verification time and rejects an authentic but causally invalid authorization. A historical predecessor path reproduced 17 frozen authorization-successor outcomes. A later in-house, instance-blind test of known case classes matched all 183 scored expectations; same-host package execution reproduced its 240 archived observations. Results support internal coherence, bounded safety, fault sensitivity, and limited constructibility, but not factual truth, general refinement, blind interoperability, field efficacy, or standards status.

Family caregivers of people living with dementia shoulder emotional and practical responsibilities, yet their own wellbeing often remains peripheral to dementia care. We built CareMirror, an envisioned caregiver wellbeing ecosystem with interconnected caregiver- and clinician-facing interfaces for longitudinal reflection, personalized support, and caregiver-controlled sharing with clinical care. We conducted semi-structured interviews with 14 caregivers, using CareMirror as a design probe to examine how they perceived this ecosystem and what expectations, concerns, and boundaries emerged around clinical connection. Caregivers valued attention to their wellbeing, longitudinal awareness, context-sensitive support, and clinical visibility when it could lead to meaningful follow-up. However, repeated reflection could become burdensome or emotionally difficult, automatic clinical sharing could inhibit candid disclosure, and participants wanted control over what information entered clinical care. They also expected AI to support reflection and communication without replacing caregiver voice or clinician judgment. We contribute design considerations for proactive, clinically connected caregiver wellbeing support.

Ambient clinical scribes (ACS) are being rapidly deployed at scale across Global South healthcare settings, aiming to reduce clinician documentation time, especially in overburdened environments like India. These ACS are primarily developed or distilled from models built and validated on Global North speech, languages and consultation styles. Indian clinical encounters are brief, triadic, multilingual, code-mixed with low-resource languages, and conducted in highly resource-constrained, noisy settings – increasing the likelihood of ASR and note-generation errors manyfold. We posit an urgent need to develop a standardized evaluation infrastructure to assess whether these systems are safe, reliable, and well-suited to the Indian healthcare setting. We substantiate our claims through a mixed-methods study – a systematic survey of publicly available patient-clinician conversational datasets, a quantitative comparison of these datasets against conversational and cultural markers drawn from the Indian clinical-communication literature, and semi-structured interviews with five organizations building and deploying ACS in India and Africa. Our survey shows that there are no publicly available, large-scale, real-world benchmarks for ACS in India, with existing datasets being overwhelmingly synthetic. We note that the available Global North datasets diverge significantly from the expected conversational and cultural structures of Indian encounters. Finally, our interviews reveal that deploying organizations have each built proprietary, incomparable evaluation pipelines, creating a fragmented ecosystem with no independent and reliable basis for procurement. We call for the development of a publicly shared, real-world, multilingual benchmark for ACS evaluation and outline the properties and policies such a benchmark would require.

The use of LLMs to complete online surveys impacts the validity of survey-based research, but detecting such usage remains underexplored. We introduce an initial benchmark dataset, namely ASURRE, for AI-assisted survey participation to capture usage strategies ranging from full generation and revision to persona-grounded agentic completion. Controlled by these strategies, LLM-assisted survey responses are generated using multiple LLMs on three real-world surveys in different disciplines, paired with genuine human responses. Our evaluation of existing machine-generated text (MGT) detectors shows that naive AI usage is readily detectable, whereas persona-grounded agents that mimic entire respondents push detector performance toward chance. We further show that agentic completion cannot fully replicate respondent-level behaviour and leaves distinctive behavioural traces. While individual cues can be circumvented by targeted prompting, a simple few-shot, training-free aggregator over these cues improves mean AUROC by +0.14 over the best existing detector across agentic settings. Our project is available at https://github.com/mike-qz-wang/ASURRE.

AI agents increasingly act through agent skills, i.e., natural-language instructions, that direct a host agent toward shell, network, credential, file, and process actions, and public registries distribute them at scale. In the first half of 2026, the OpenClaw AI agent went viral, and its public skill registry boomed: the observable stock nearly doubled in 91 days, and a majority of the listings visible in June were created in just two months. By the end of our study window, the wave had crested, and monthly listing creation and core-repository activity were falling from their spring peaks. This paper measures what the boom left behind, drawing on the OpenClaw Git history, its GitHub issues and pull requests, and three ClawHub registry snapshots. Attention is concentrated: the top 10% of skills received 46.93% of all downloads. No simple skill features (like size or download counts) remained a stable predictor of continued listing once creation cohort and skill age were controlled. Human scrutiny did not stay: 77.86% have zero stars and zero comments, while 85.06% of the readable skills carry privilege evidence. And automated cleanup is not ready: the three security scanners disagreed on 23,702 of the 61,990 skills they all cover. After human adjudication, weighted scanner sensitivity against the reference standard ranged from 21.67% to 61.06%. Governing fast-growing agent-skill registries cannot rely on simple metadata or single scanner scores; it requires robust, transparent measurement and independent validation.

Medical AI models hold immense potential to improve patient outcomes, but they are also known to unintentionally memorise individual records from their training datasets. While such memorisation has been linked to targeted privacy attacks, its consequences for clinical deployment, where patients may be assessed by a model that saw their historical data during training, remain poorly understood. Here we show that predictions on a patient's unseen future data can change significantly if a model observed that same patient's anonymised historical data during training, a phenomenon we term "memorisation bias". We demonstrate that this bias exists across diverse data modalities and model architectures, and over prolonged time spans: in some cases, memorisation bias persists on future records acquired decades after the historical records used for training. Moreover, in simulated prospective deployment, memorisation bias has asymmetric effects on the diagnostic accuracy of returning data contributors. When a patient returned with a de novo condition absent from their historical records in the training dataset, diagnostic sensitivity decreased significantly compared to an otherwise identical model not trained on their historical data. Conversely, when their health state was unchanged, both sensitivity and specificity were significantly inflated. Our findings reveal a previously uncharacterised risk in medical AI that arises when a model is deployed on patients who contributed to its training data. This exposes a shortcoming of current model development practice: the de-identification measures designed to protect patients' privacy make it difficult to identify returning contributors and exclude them from the AI-assisted interpretation of their own future data. Mitigating memorisation risks may thus require changes to current model training and deployment protocols.

Modelling with mathematical formalisms like logical formulas, mathematical equations, or regular expressions is an important yet challenging task for students of computer science and other STEM disciplines. Identifying common mistakes occurring in this context is an important step towards helping struggling students by providing targeted high-quality feedback, e.g. in interactive learning systems. We present a tool-supported workflow that allows to (1) identify candidates for common mistakes that explain many student mistakes in large educational data sets, (2) cluster candidates according to similarities, and (3) visualize resulting clusters for instructors and CS education researchers. The visualization is designed to help researchers to identify common modelling mistakes. The candidates for common mistakes are represented by bug fixing transformations that translate incorrect formalizations into correct formalizations; they are generated by an LLM and validated algorithmically. We show that this approach works well by reproducing common mistakes in propositional logic modelling that were identified by hand in the literature; showing that, unlike other algorithmic approaches, the LLM-based approach is suitable for very large sets of data; and applying it to multiple other formalisms to showcase it generalizes beyond propositional logic.

Large Language Models are now common in student assessment, but we know little about how student demographics affect their use. Sometimes, considering student demographics may be necessary – for example, to improve readability for users with lower educational levels. However, it also risks being a cause of discrimination, e.g., when assigning lower scores to students from lower socioeconomic backgrounds. We set up controlled prompts to test 1) explicit demographic effects, where we mention demographic details directly, and 2) implicit effects, where we use conversation history as a demographic signal. We test these settings in three tasks: Automated Essay Scoring, Formative Feedback, and Metalinguistic Question Answering. We test six state-of-the-art LLMs on these tasks. In both explicit and implicit cases, the models pick up on demographic cues and can change their scoring, feedback, and answers accordingly. We find that LLMs frequently adjust the readability of feedback to education levels when these are explicitly mentioned. On the other hand, implicit conditions produce unpredictable biases, such as in question answering, where responses from lower-education levels receive lower sentiment scores. Our results provide clear evidence of demographic sensitivity in LLMs for educational assessment tasks.

AI companions can provide meaningful relationships, yet these relationships remain vulnerable to platform-initiated changes. We study AI companion disruptions: platform changes that alter or terminate users' ongoing companionship with an AI. We compile 30 disruption events across major platforms, develop a taxonomy of six disruption types, identify three broad reasons for disruption, and propose a risk-assessment framework comprising four dimensions: relational discontinuity, population vulnerability, communication deficit, and transition-support deficit. Using longitudinal Reddit data, we estimate community-level psychosocial responses with a hierarchical Bayesian interrupted time-series model incorporating predictive controls. Across events, disruption onset was associated with immediate increases in anxiety, stress, suicidal expression, and grief activation, with relational discontinuity and transition-support deficit being associated with more adverse immediate responses across several outcomes. Our findings provide a cross-platform characterization of AI companion disruptions, quantitative evidence of their psychosocial impacts, and a prospective framework for assessing their potential risks before implementation.

Generative AI has become a routine resource in programming education, and most institutional responses to it are attempts at control, either by restricting access or by offering students a controlled version of the technology. This paper reports a seven-week mixed-methods pilot study in a master's-level data analytics course, in which 33 students were randomly assigned either to a scaffolded AI Study Coach embedded in the notebook-based laboratory sessions or to unrestricted use of AI tools of their own choosing. The Coach offered stepwise hints, did not generate code, limited the number of hints per session, and required a short reflection at the end of each session. The design assumed, in line with scaffolding theory and recent experimental evidence, that guided and limited support would build confidence and reduce over-reliance, and that the scaffolded group would learn more. Assignment performance did not differ between the conditions. Students in the Coach condition reported higher confidence but managed the hint budget poorly, while students in the unrestricted condition were satisfied with their tools and uneasy about how much they depended on them. In interviews, students in both conditions identified awareness of their own reliance on AI as the most valuable outcome of the course. Students who had formulated their own rules for when to use AI performed better in both conditions, and those with the best understanding of how the models work, in every case self-taught, used the tools most deliberately and achieved the highest scores. The design of the tool mattered less than the students' capacity to govern their own use of it, a capacity that is at present acquired by chance. The paper argues that the appropriate response is structural: assessment that grades the reasoning behind AI-assisted work, and AI literacy taught explicitly as a core skill.

Technological advancements are enabling increasingly systematic and large-scale data collection across all areas of science, driving scientific innovation. In particular, AI research exemplifies this trend, having advanced rapidly through the assembly of massive datasets used to train and evaluate machine learning models. However, the escalating demand for data, the difficulty of creating high-quality datasets, and the exhaustion of easily accessible data sources in AI research raise important questions about how to maximize the value of existing datasets through recombination and repurposing. Here, we draw on two theoretical frameworks—recombinational novelty and transformational creativity—to examine the practice of data repurposing and its scientific impact. Focusing on AI, we analyze scientific outcomes associated with data repurposing across more than 10,000 machine learning papers. First, we find that although most repurposed datasets do not achieve broad visibility in the short term, data repurposing is associated with greater disruption. Second, when repurposed data is adopted by subsequent research, the repurposing paper is associated with higher disruption and increased citation impact. Third, repurposing teams tend to be more experienced, more institutionally prestigious, and involve academic–industry collaboration. However, team characteristics poorly predict which repurposed datasets will be adopted by the community. These findings suggest that data repurposing may be an important approach to scientific discovery, and that its successful adoption is more common among larger teams and collaborations spanning academia and industry.

Public evidence can become sufficient to settle a prediction-market contract before the venue records its first determination, but the relevant boundary depends on the applicable rule version, exact release object, source hierarchy, correction history, and unfinished contract conditions. This paper defines two Kalshi clocks: first decidability, the earliest contemporaneous singleton in the rule-evidence mapping, and stable decidability, the retrospective earliest time after which the same singleton remains unchanged through finalization. A completed retrospective identification test establishes a narrow feasibility result. In a frozen blind pilot, all 25 identities and blinding checks passed and current rule text was recovered for all 25; no exact or bounded historical rule version and no exact or bounded official source-release object was recovered. The full historical recovery covered 152,694 ordinary tickers, 11,530 exact event identities, and 6,540 read-only official requests, with zero historically eligible events and zero historically eligible tickers. This is an observability result, not a claim that no market was decidable or that public evidence never existed. A completed prospective infrastructure shakedown established observation capability for three source programmes across 25 markets, with integrity revalidation of 781,266 lifecycle frames, 22 closed lower-bounded reconnect receipts, no unresolved reconnect gap, no due-but-missed official release, and an inactive price layer. Production evidence enrollment is active. The prospective sample is constructed only at enrollment close from prospectively frozen identities and pre-outcome fields; its frozen target size is selected mechanically under the registered full, reduced, exploratory, or no-go support disposition. No contractual-decidability clock, human-adjudication, price, or cross-venue result is reported here.

Quantifying organizational environmental action from publicly available web content remains a challenging environmental data science problem because relevant information can be dispersed across multiple webpages and is primarily communicated through unstructured text. We present a scalable computational framework for transforming organizational web content into structured measures of environmental action and demonstrate the approach using Jewish congregations in the United States. We constructed a national database of 4,964 congregations by integrating multiple geospatial, knowledge-base, directory, and manually reviewed sources. Of these, 2,657 had active websites that were successfully crawled, producing a corpus of 154,454 webpages. We compared three approaches for detecting environmental actions: keyword retrieval followed by large language model (LLM) classification, semantic vector retrieval followed by LLM classification, and direct LLM classification classification without preliminary retrieval. Agreement with an expert human reviewer was lowest for keyword retrieval ($\kappa$ = 0.26), higher for semantic vector retrieval ($\kappa$ = 0.42), and similar for direct LLM classification ($\kappa$ = 0.40). Although semantic retrieval achieved the highest agreement, its retrieval recall was 0.87, indicating loss of relevant content before classification. Applied to the complete corpus, direct LLM classification identified at least one environmental action at 1,398 congregations (53%), providing greater coverage than either retrieval-based approach. These results demonstrate that preliminary retrieval can reduce computational cost but may exclude relevant information before it reaches the classifier. The framework provides a reproducible approach for extracting organization-level environmental information from unstructured web content that can be adapted to other institutions.

This paper presents an end-to-end approach for generating context-specific large language model (LLM) benchmark datasets by combining expert input with synthetic data generation. Existing benchmark construction methods often trade off validity and scalability: datasets designed with domain experts can produce high-quality evaluations but are slow and costly to create, while synthetically generating data may scale efficiently but often results in unrealistic, redundant, or out-of-scope examples. To address this gap, we introduce a schema eliciting key information about the goals, scope, and context of an evaluation task, and use this information to guide synthetic data generation. We further define four criteria grounded in measurement validity for assessing dataset quality: coverage, diversity, content realism, and stylistic realism. Using these criteria, we show how expert-informed scaffolds can guide synthetic data generation toward more valid benchmarks. Through quantitative evaluations and a real-world case study with domain experts, we demonstrate that our approach improves benchmark data quality over existing methods while preserving validity. We additionally analyze how different types of schema information affect different dataset quality criteria, and provide practical guidance on which information to prioritize collecting under resource constraints.