Named after the hundred-eyed watchman of Greek myth, Argus watches the education landscape: spotting new opportunities, pressure-testing the ventures we're building, and tracing every read back to the real-world signals behind it.
The evidence library: the raw signals the pipeline is watching across the education ecosystem. Every idea is built from these.
arXiv:2607.09306v3 Announce Type: replace-cross Abstract: Behavioural auditing asks whether a language model behaves as it claims, but detection scores are reported without separating two targets: whether a reply was produced under a behaviour-inducing condition (exposure) and whether the behaviour surfaced in it (manifestation). Scoring a compact 146-million-parameter auditor's frozen-representation read-out and a frontier judge against each label on the identical 720 replies, the gap between the instruments moves by roughly 0.2 AUROC when the target changes. Under the judge's deployed interface, a single verdict, the ranking reverses: the auditor leads on exposure, 0.804 against 0.718, and trails on manifestation, 0.690 against 0.811. Matching the output resolution from either direction, by asking the judge a target-specific question answered with a continuous confidence score or by thresholding the auditor's read-out, removes the reversal but not the interaction, which excludes zero
arXiv:2603.26487v2 Announce Type: replace-cross Abstract: Generative AI (GenAI) is playing an increasingly important role in open source software (OSS). Beyond completing code and documentation, GenAI is increasingly involved in issues, pull requests, code reviews, and security reports. Yet, cheaper generation does not mean cheaper review - and the resulting maintenance burden has pushed OSS projects to experiment with GenAI-specific rules in contribution guidelines, security policies, and repository instructions, even including a total ban on AI-assisted contributions. However, governing GenAI in OSS is far more than a ban-or-not question. The responses remain scattered, with neither a shared governance framework in practice nor a systematic understanding in research. Therefore, in this paper, we conduct a multi-stage analysis on various qualitative materials related to GenAI governance retrieved from 67 highly visible OSS projects. Our analysis identifies recurring concerns across co
arXiv:2607.20835v2 Announce Type: replace Abstract: Audio description (AD) makes film and television accessible to blind and low-vision (BLV) audiences by narrating characters' actions. However, in scenes with lots of dialogue, AD often omits important actions because it is constrained not to overlap with speech. It is not yet known how to convey characters' actions during dialogue. We present Sonic Stage, a system that transforms dialogue videos into interactive spatial soundscapes, enabling BLV audiences to intuitively understand characters' actions and movements through immersive auditory cues. Sonic Stage conveys essential visual information during dialogue through three auditory techniques: (1) spatialized dialogue to represent spatial layout, (2) diegetic sound to convey character actions, and (3) interactive descriptions to provide context-specific visual details. Evaluation with 12 BLV viewers showed that Sonic Stage significantly improved video comprehension, spatial presence,
arXiv:2606.04308v2 Announce Type: replace Abstract: Reading augmentation systems increasingly help readers process text at scale. While these tools address real constraints of time and cognitive load, they often implicitly frame reading as information transmission, or "reading to discard," delegating interpretation and effort to the machine. Yet this delegation changes the outcome of reading. For example, in scholarly reading, deciding what a research text implies and why it matters is central to the work of scholarly production. We propose creative reading as an alternative goal: reading augmentation that supports readers in creating both readings and themselves as readers. By putting literary and narrative theories into conversation with scholarly sensemaking and creativity support, we present a provocation-oriented design space for valuing the process of reading as a way of preserving a plurality of readings and transforming readers over time.
arXiv:2604.23749v2 Announce Type: replace Abstract: Real-world environments evolve continuously, yet blind and low-vision (BLV) individuals often have limited access to understanding how they change over time. Unexpected or relocated objects, layout modifications, and content updates (e.g., price changes) can introduce safety risks and cognitive burden. While existing visual assistive technologies can describe immediate surroundings, they operate as one-off interactions and lack mechanisms to surface meaningful changes across revisits. Informed by a survey of 33 BLV individuals, we develop StateScribe, a system that supports accessible awareness of real-world changes across revisits. StateScribe employs a dual-layer memory architecture that integrates episodic scene memory and object-centric temporal memory to enable scalable and structured change tracking. It provides both live descriptions of the current scene, and descriptions of what has changed, when and where it occurred across r
arXiv:2510.26172v2 Announce Type: replace Abstract: Social media platforms generate volumes of heterogeneous data, capturing user behaviors, textual content, and network structures. Analyzing such data is crucial for understanding phenomena such as opinion dynamics, community formation, and information diffusion. However, discovering insights from this complex landscape is exploratory, conceptually challenging, and requires expertise in social media mining and visualization. Existing automated approaches, including large language models (LLMs), remain largely confined to structured tabular data and cannot adequately address the heterogeneity of social media analysis. We present SIA (Social Insight Agents), an LLM agent system that links heterogeneous multi-modal data, including raw inputs (e.g., text, network, and behavioral data), mined analytical results, and rendered visual artifacts, through coordinated agent flows. Guided by an insight-oriented taxonomy connecting insight types wi
arXiv:2501.11840v2 Announce Type: replace Abstract: Systematic reviews are time-consuming endeavors that require knowledgeable human reviewers to screen studies for relevance and extract data following a specific coding scheme before any analysis or synthesis can occur. Large language models (LLMs) hold promise for substantially accelerating this process and reducing reviewer workload, yet their application within the context of systematic reviews in the field of education remains underexplored. We address this issue in two ways: through empirical studies and the iterative development of an open-source software tool. First, we conducted two empirical studies examining the efficacy of using LLMs for data extraction using data from a published review on pedagogical agents. We extracted a variety of data types from 112 studies and compared the results to data extracted by human coding. Results indicate that LLMs struggled with extracting data accurately and therefore are not ready to be u
arXiv:2402.17838v2 Announce Type: replace Abstract: In this study, we explore the effectiveness of persuasive messages endorsing the adoption of a privacy protection technology (IoT Inspector) tailored to individuals' regulatory focus (promotion or prevention). We explore if and how regulatory fit (i.e., tuning the goal-pursuit mechanism to individuals' internal regulatory focus) can increase persuasion and adoption. We conducted a between-subject experiment (N = 236) presenting participants with the IoT Inspector in gain ("Privacy Enhancing Technology" -- PET) or loss ("Privacy Preserving Technology" -- PPT) framing. Results show that the effect of regulatory fit on adoption is mediated by trust and privacy calculus processes: prevention-focused users who read the PPT message trust the tool more. Furthermore, privacy calculus favors using the tool when promotion-focused individuals read the PET message. We discuss the contribution of understanding the cognitive mechanisms behind regul
arXiv:2607.28302v1 Announce Type: cross Abstract: For empirical studies, social and health scientists give background characteristics of their sample and summarize them in the famous "Table 1". When treatment/ control groups are present, this table gives summary statistics by group to see whether the background characteristics differ by group. We propose snapshot plots --- parallel univariate plots with consistent highlighting --- to visualize such tables. Compared to "Table 1", such plots are designed to facilitate comparisons of background characteristics --- in particular among groups --- and give more detail on numerical variables. We provide a web app as well as a python implementation of snapshot plots. Snapshot plots arise as edge cases of hammock plots (parallel coordinate plots for mixed categorical/ numerical data). We demonstrate the usefulness of snapshot plots for two "Table 1"s.
arXiv:2607.27851v1 Announce Type: cross Abstract: Emotional dialogue research includes two influential strategy traditions. Empathetic dialogue prioritizes understanding a speaker's emotional experience. Emotional support conversation selects and sequences support for the seeker's current needs. Sustained use introduces a further goal. Effective support should sustain users' capacities for emotion regulation, coping, self-endorsed decisions, and social connection across the interaction lifecycle. We propose capability-sustaining emotional dialogue (CSED) as a longitudinal research paradigm that aligns supportive strategy with this goal and organizes data, models, system design, evaluation, and governance around repeated use, non-use, transition, and termination. A targeted literature-and-corpus audit motivates this position. In a PRISMA-ScR-guided sample, 95% of 60 system-building papers pursue relief-oriented goals. None evaluates capability or longitudinal outcomes, and only 1 consid
arXiv:2607.27739v1 Announce Type: cross Abstract: Context compression discards most of a document before a language model reads it, and is normally evaluated by downstream task accuracy - which makes another model the judge of what mattered. Naturalistic social highlighting offers a non-circular reference: many people independently marking passages on the same page. But the obvious metric, the fraction of crowd-marked sentences a compressor keeps, is confounded twice: crowd marks are front-loaded and crowd-marked sentences are longer, so any method favouring early or long sentences scores well regardless of readers. We remove both by matching each marked sentence against unmarked sentences of the same document at equal relative depth and equal within-document length rank, and we calibrate every estimator on synthetic nulls built from position and length alone - a step that matters, since depth-only stratification returns a false positive on 20-36% of nulls containing no effect. On 120
arXiv:2607.27568v1 Announce Type: cross Abstract: Recognition accuracy obtained during a recording session does not persist when a user puts on the electrodes again after the electrodes had previously been removed. The electrodes may have moved slightly, the skin may be drier or wetter, or the elbow may be positioned differently; these factors all contribute to day-to-day variability and therefore represent a major obstacle to implementing successful pattern-recognition based myoelectric control systems in daily practice. However, simply recalibrating a user's hand for 20 min at every doff/don event is a clearly unrealistic expectation. A montage-agnostic encoder built for cross-user, cross-montage transfer is trained here using data collected during a particular recording session, and then applied to data collected later in a different recording session without adjusting anything, on the ten intact subjects of NinaPro DB6. The performance of this approach is compared to that of a per-
arXiv:2607.27565v1 Announce Type: cross Abstract: Pattern-recognition control promises a myoelectric prosthesis that responds to many intended gestures rather than one or two, but the promise has stayed in the laboratory. A recogniser trained on one person rarely transfers to the next, and useful performance usually demands a fresh round of labelled calibration from the end user. A montage-agnostic encoder is introduced that reads each electrode with shared weights and locates it by its physical coordinate rather than its index, so one architecture ingests any channel count without montage-specific parameters. Trained across users, it exceeds a per-user Hudgins and linear-discriminant classifier by 0.234 macro-F1 on DB1 for every held-out subject and by 0.108 on DB2, and falls below it on the ten-subject DB5. Each of the encoder's three key components individually accounts for more than half of its 3-shot macro F1 in an otherwise budget-matched ablation study. A controlled subject-coun
arXiv:2607.27463v1 Announce Type: cross Abstract: Dimensionality Reduction (DR) is a fundamental tool for high-dimensional data exploration, reducing the complexity of latent spaces of machine learning models, and assisting in the explanation of complex opaque models. However, non-linear DR techniques often function as opaque transformations themselves, making it challenging to understand how individual features influence instance positioning in the reduced space. This lack of transparency complicates the analysis and interpretation of structural patterns, hindering the ability to reason about the organization of high-dimensional data based on the projected layout. In order to address this challenge, dimensionality reduction explanation methods have shown promise in improving the understanding of the observed groups and cluster structures. Unfortunately, existing DR explanation approaches tend to suffer from limitations such as multiple attributions per feature and restricted applicabi
arXiv:2607.28583v1 Announce Type: new Abstract: Cross-reality collaboration increasingly connects immersive and desktop users within synchronized workspaces, yet little is known about how bidirectional projection techniques between immersive 3D layouts and desktop 2D views influence communication. Spatial referencing depends on shared spatial understanding, but different mappings preserve and distort geometric relationships in different ways, altering perceived adjacency, orientation, and coverage across collaborators' views. We present CrossAtlas, a synchronized PC-VR collaboration platform that integrates multiple bidirectional projection techniques, including three planar projection variants and equirectangular, a spherical projection variant, across layouts of varying curvature. In a controlled study with 24 dyads, collaborators completed spatial referencing tasks under different projection-layout conditions while we collected performance and subjective measures. Our results show t
arXiv:2607.28552v1 Announce Type: new Abstract: Understanding automated vehicles (AVs) is crucial to improving their acceptance. Numerous approaches to visualizing relevant traffic information to passengers have been proposed and empirically evaluated. As this is time-consuming, costly, and reduces the possible design parameters, we employed multi-objective Bayesian optimization to optimize the design of visualizations in AVs. In particular, we evaluated multi-session aspects involving iterative optimization. We optimized the design for passenger trust and perceived safety while minimizing cognitive load. Results from an online study (N=74) show that this method effectively identifies visualization design parameter values that improve trust, safety, and predictability while making the design process more efficient and scalable. However, shortcomings of the computational approach when optimizing for subjective measurements are highlighted and discussed.
arXiv:2607.28544v1 Announce Type: new Abstract: Automated vehicles promise to improve accessibility and access to personal mobility for everyone. However, their design and current research trends in visualizing relevant information do not reflect this commitment to accessibility for users with visual impairments. Therefore, we designed and implemented a visual and auditory communication concept for people with visual impairments seated inside fully automated vehicles. Furthermore, in an online video-based study (N=35, 12 with visual impairments), we compared three levels of auditory information communication: low (safety-relevant information only), medium (additionally including vehicle control and route updates), and high (additionally including sightseeing and destination information). Results showed that trust and user experience significantly improved with additional information, with a corresponding, albeit less robust, effect on perceived safety. However, they also revealed that
arXiv:2607.28239v1 Announce Type: new Abstract: Given the profound societal impact of vaccine-skeptical content on social media, community-driven counterspeech has emerged as a promising participatory response to contest and curb such objectionable content. Yet crafting effective counterspeech remains challenging for ordinary users, limiting their willingness and ability to engage constructively. We designed and evaluated three generative AI-assisted counterspeech writing systems that vary by assistance stage (co-writing vs. re-writing) and mode (guided vs. unguided) to support lay users' responses to vaccine-skeptical content. We ask whether AI can help users craft counterspeech perceived as both effective and authentic, which forms of AI support work best, and through what mechanisms. In a randomized controlled trial with social media users, participants wrote counterspeech responses to both statistical and narrative vaccine-skeptical content. Across evidence types, AI-assisted writi
arXiv:2607.28204v1 Announce Type: new Abstract: Continuous emotional arousal quantification remains bottlenecked by time-consuming and labor-intensive manual annotation. This work investigates group-level EEG dynamic neural synchrony (DNS) as a principled signal for continuous arousal quantification that bypasses per-subject manual labeling. Using Correlated Component Analysis (CorrCA) with sliding-window computation across four EEG datasets spanning 142 subjects and over 207 hours, we systematically evaluate DNS as a group-level marker for emotional arousal dynamics. Three key findings emerge. First, DNS exhibits significant emotion information from valence-dependent differences (all p<0.003), with positive emotions eliciting higher synchrony. Second, DNS correlates more strongly with the first-order derivative of arousal than with raw arousal values, revealing that neural synchrony captures the rate of emotional change rather than static intensity. Third, we provide the first systema
arXiv:2607.28203v1 Announce Type: new Abstract: Students differ in how they prefer to engage with learning resources, with some favoring textual materials and others visual or video-based content. Recent advances in generative AI have led CS education research to focus on text-based AI tools for developing learning resources. However, advances in AI video models and the rapid proliferation of AI video generation tools have made it possible for instructors to create high-quality personalized educational videos efficiently and cost-effectively. Understanding students' perceptions of AI-generated videos is thus critical for helping CS instructors know when and how to use them purposefully. To address this gap, we conducted a descriptive post-test survey study in which 170 computing students at two U.S. institutions watched three 3-minute AI videos on the Markdown markup language created with Knowlify. Students then completed a survey about their perceptions of the Markdown videos and thei
arXiv:2607.28155v1 Announce Type: new Abstract: Background. New forms of music writing using computers have arisen in the past 20 years. Most of them use the capacities of digital manipulation of data like animation, algorithmic processing, cinematic view, and much more. These scores use dynamic musicography, and all of them share a problem. They have readability problems. We will argue that this problem can be addressed by mathematical tools. Aims. Take the Autopsias [Autopsies] meta-compositional system as a study case for starting the construction of a mathematical framework that can overview the readability for cynetic musicography. The Autopsias system is the process of transforming a musical score dynamically. Methodology. We will start to build a mathematical framework by taking a group of basic topological concepts, and applying them after a bridge between a printed score and a dynamic computational re-appropriation of it has taken place. We will observe the writing and the per
arXiv:2607.28072v1 Announce Type: new Abstract: Commuters often encounter crowding in railway systems, particularly in queues where passenger density varies throughout the day. This introduces uncertainty in crowdedness, making it difficult for individuals to anticipate conditions and plan their trips effectively. Crowdsourcing has been a valuable method for collecting localized user data. But the unpredictability of crowds and the uncertainty of crowdsourced information pose new challenges for decision-making. However, we know little about how to effectively visualize uncertainty in crowdedness to support informed commuting decisions, particularly when using crowdsourced ordinal data. Here, we investigated different uncertainty visualizations and their effectiveness in representing the variability and reliability of crowdsourced crowding data. They were evaluated through an online study, and we found that cluster visualization is best suited to reduce cognitive load while maximizing u
arXiv:2607.27938v1 Announce Type: new Abstract: Composite visualizations integrate multiple visualizations to represent complex datasets effectively, but their intrinsic composite designs often impose a high initial cognitive load on novice users. Existing visualization onboarding approaches are typically platform-dependent, require substantial manual authoring effort, and struggle with the structural complexity of composite visualizations, limiting their general applicability. We present VizPilot, an automated visualization onboarding approach that reverse-engineers composite visualization structure to generate interactive onboarding experiences directly from raw visualization artifacts. VizPilot consists of two modules: a Composite Visualization Analyzer and an Onboarding Interface. Leveraging Multimodal Large Language Models (MLLMs), the Analyzer employs a two-stage pipeline that decomposes a visualization into visual components, extracts structured explanations, and maps them to pr
arXiv:2607.27863v1 Announce Type: new Abstract: Computer Science students in introductory programming courses struggle with emotional challenges such as low levels of sustained interest and a lack of self-belief. However, students are typically introduced to disciplinary strategies (such as debugging strategies) and not the metacognitive nor self-regulation strategies that could help them overcome such emotional challenges. This paper addresses this issue by introducing creative task cards designed to support students' reflection, self-efficacy, and self-regulation in programming assignments. To evaluate the cards, twenty-nine students used them to complete a mock coding assignment in-class and participated in surveys and focus groups. Our initial qualitative insights suggest that students: found value in taking breaks, particularly when they could leave the classroom to reflect with peers; could use drawing to better reflect on feelings of cognitive overload; and felt more capable pro
arXiv:2607.27697v1 Announce Type: new Abstract: Immersive environments, e.g., virtual reality (VR), offer a unique approach to exploring complex 3D datasets, where data is often heavily occluded and exploration incurs a high cognitive load. We propose DP-LENS, a density-aware polyfocal fisheye lens equipped with topology-driven auto-routing. While preserving peripheral context through geometric deformation and 3D perspective techniques, it enables users to explore 3D data with a lower cognitive load. To facilitate hands-free macro-navigation, we integrate a Large Language Model (LLM) to serve as a supplementary voice-based target selection tool that initiates the auto-routing algorithm. Two user studies with 34 participants investigate the potential benefits of this system. Our first study (N=18) compared the manual DP-LENS against two industry-standard baselines (i.e., World-in-Miniature and volumetric slicing) in heavily occluded 3D datasets. The results show that DP-LENS significant
arXiv:2607.27369v1 Announce Type: new Abstract: State-of-the-art head-mounted displays (HMDs) enable gaze-based selection in virtual environments. Yet, these HMDs suffer from the vergence-accommodation conflict (VAC), which is known to affect interaction performance. The VAC might influence gaze-based selection performance because it directly affects eye-movement behavior. Thus, in this paper, we investigate how the VAC influences gaze-based 3D target selection across varying depth conditions. Our results show that as the (visual) depth increases, user performance significantly decreases with gaze-based selection. Moreover, a previously suggested Variation in Diopter Fitts' law model captured this performance change better relative to a linear model. These findings provide evidence that gaze-based pointing is negatively affected by the VAC and highlight the importance of accounting for depth-dependent factors when designing gaze-based interaction in 3D environments.
arXiv:2607.27212v1 Announce Type: new Abstract: Children with Autism Spectrum Disorder in Arabic-speaking countries face compounded barriers to effective speech and language therapy: a shortage of qualified specialists, limited service reach beyond urban centers, and a near-total absence of culturally grounded digital therapy materials. We present Digital Harf, a pervasive, multimodal AI platform that extends clinician-led speech and language therapy into the home. The system integrates three therapeutic modules - language therapy, speech intelligibility, and picture description - within a unified workflow that adapts to each child's performance over time. To address the critical shortage of Arabic therapy content, we introduce an Agentic Synthetic Data Engine (ASDE) that automatically generates culturally relevant images, prompts, and language tasks guided by explicit therapeutic and cultural criteria. Expert evaluation with 13 licensed Speech-Language Pathologists yielded a 90.1% cli
arXiv:2605.11640v2 Announce Type: replace-cross Abstract: This paper studies behavioral concentration in Polymarket's public executed-fill record and formalizes what that record can and cannot identify. A pre-publication reconciliation corrects the empirical scope: the archived extraction covers the legacy CTF Exchange over Polygon blocks 86,008,447-86,107,178, approximately 25 April 2026 17:09 UTC through 28 April 2026 00:00 UTC, rather than the full 21-27 April week stated previously. It contains 13,356,931 OrderFilled records, 77,204 addresses with at least five attributed records, and 43,116 token identifiers; negative-risk markets are absent. The archived feature construction credits both maker and taker addresses on each record. This convention is not invariant to match fragmentation, and mint/burn executions do not admit a universal buyer/seller interpretation. The reported one-cluster result is therefore retained only as a null under the original record-level representation, no
arXiv:2512.15011v3 Announce Type: replace-cross Abstract: Artificial intelligence (AI) increasingly generates the very content used to train future AI systems. This feedback loop can degrade model quality, reduce informational diversity, and ultimately drive knowledge collapse, i.e. a degradation to a narrow and inaccurate set of ideas. We ask: to mitigate collapse, is it better to concentrate the internet's knowledge into a handful of dominant models (referred to as an AI monoculture), or to distribute it across a diverse ecosystem of models? To study the effect of diversity on model performance, we randomly segment the fixed training data across an increasing number of language models and evaluate the resulting ecosystems of models over ten self-training iterations. Our results show that diversity improves long-term performance of models, while monoculture accelerates collapse. Specifically, we observe that the optimal diversity level (i.e., the level that maximizes performance) incr
arXiv:2509.12290v3 Announce Type: replace-cross Abstract: Human oversight of AI is promoted as a safeguard against risks such as inaccurate outputs, system malfunctions, or violations of fundamental rights, and is mandated in regulation like the European AI Act. Yet debates on human oversight have largely focused on its effectiveness, while overlooking a critical dimension: the security of human oversight. We argue that human oversight creates a new attack surface within the safety, security, and accountability architecture of AI operations. Drawing on cybersecurity perspectives, we model human oversight as an IT application for the purpose of systematic threat modeling of the human oversight process. Threat modeling allows us to identify security risks within human oversight and points towards possible mitigation strategies. Our contributions are: (1) introducing a security perspective on human oversight, (2) offering researchers and practitioners guidance on how to approach their hum
arXiv:2505.12507v2 Announce Type: replace-cross Abstract: Despite the success of machine-generated text detectors, the black-box nature remains a critical limitation. Traditional explainability methods rely on token-level saliency, insufficient to reveal the high-order structural dependencies that distinguish LLM outputs. In this paper, we propose \textsc{LM$^2$otifs}, a principled framework that shifts detection from linear sequences to graph-structured manifolds. We first provide a theoretical grounding based on probabilistic graphical models, demonstrating that detection performance is more distinguishable in the graph-topological space. Driven by this theory, \textsc{LM$^2$otifs} transforms text into lexical co-occurrence graphs to preserve latent structural fingerprints. The framework employs Graph Neural Networks for robust detection and utilizes graph-specific explainers to extract interpretable motifs. Crucially, our experiments reveal that these structural motifs achieve highe
arXiv:2606.01171v2 Announce Type: replace Abstract: Artificial intelligence (AI) can reproduce and amplify the structural inequities faced by minoritized communities. Participatory AI has been proposed as a response, but participation typically starts after problem definitions and success criteria have been set, leaving limited room for minoritized communities to reshape what an AI system is for. We propose AI From the Margins (AIM): a methodological stance that articulates the conditions under which lived experiences of minoritized communities can be elicited, centered, and carried forward to inform participatory AI design. AIM is not a fixed protocol; it articulates a set of preconditions that can be enacted through different techniques in different settings. We applied AIM in a Dutch healthcare context in eight sessions with 13 women and non-binary people of color and five municipal policy workers, namely through (1) narrative elicitation using the Biographic Narrative Interpretive
arXiv:2512.06354v2 Announce Type: replace Abstract: This paper addresses a critical gap in the risk assessment of AI-enabled safety-critical systems. While these systems, where AI systems assist human operators, function as complex socio-technical systems, existing risk evaluation methods fail to account for the associated complex interaction between human, technical, and organizational components. Through a comparative analysis of system attributes from both socio-technical and AI-enabled systems and a review of current risk evaluation methods, we confirm the absence of explicit socio-technical considerations in standard risk expressions. To bridge this gap, we introduce a novel socio-technical alignment ($STA$) variable designed to be integrated into the traditional risk equation. This variable estimates the degree of harmonious interaction between the AI systems, human operators, and organizational processes. A case study on an AI-enabled liquid hydrogen ($LH_2$) bunkering system de
arXiv:2607.28617v1 Announce Type: cross Abstract: System prompts are instructions configured by developers to govern the behaviors of foundation models in AI applications. They are used throughout commercial AI products, but are rarely disclosed to the public or regulators, creating a serious trust and accountability gap in the wide deployment of AI systems. In this paper, we introduce Artificial Intelligence System Prompt Assurance (AISPA), a user-centric framework for systematically auditing system prompts in AI systems. AISPA examines specific parts of a system prompt and evaluates them along eight dimensions that matter to users. We then use this framework to review 3,249 instructions from system prompts in 88 commercial AI products, classifying each instruction as either protective (of users) or problematic. Our audit surfaces four core findings. First, system prompt design varies substantially across products and developers, with some organizations averaging over 60 protective in
arXiv:2607.28497v1 Announce Type: cross Abstract: Algorithmic recourse aims to provide individuals with actionable changes to improve their predicted outcomes in high-stakes classification settings, such as loan and mortgage applications. However, most existing approaches focus only on flipping a model's prediction, without accounting for whether the recommended changes lead to genuine improvement in an individual's true qualifications or merely enable strategic gaming of the classifier. Consequently, deployed recourse policies can induce behavioral responses that degrade predictive accuracy and become ineffective after model retraining. In this work, we formalize this failure mode through a causal performative framework for recourse. We model how recourse actions propagate through a structural causal model, capturing interactions among features as well as their effect on the true label. These causal responses induce a non-convex optimization problem, even under standard convex losses.
arXiv:2607.28319v1 Announce Type: cross Abstract: This work presents Fairness Pruning, a lightweight structural intervention method designed for the management and future mitigation of demographic bias in large language models (LLMs). As a foundational empirical validation of this method, this work focuses on causal bias localization. Using minimally contrastive prompt pairs and inference-time activation capture, the method identifies neurons that react differentially when processing demographic attributes in GLU architectures, evaluating the signal at the down_proj input. Empirical evaluation was conducted on models of up to 3 billion parameters (Llama-3.2 family and Salamandra-2B), combining standardized benchmark evaluation with qualitative text generation experiments. Results demonstrate that zeroing the identified neurons alters how the model responds to associated demographic variables. However, rather than producing flat mitigation, the intervention causes bidirectional bias des
arXiv:2607.28128v1 Announce Type: cross Abstract: LLM tutoring poses a measurement problem: can a general-purpose helpfulness rubric distinguish direct answer-giving from pedagogical guidance? We audit this signal in a pre-registered study. Within each of three tutor bases, we compare conversational and pedagogical policies instantiated with the same underlying model and paired with one fixed weak simulated student. Deterministic detectors measure answer leakage and next-turn independent work. Claude Opus 4.8 is the frozen, condition-blind primary judge. After the Opus scores were fixed, GPT-5.6 Sol was prospectively specified for a post hoc robustness audit of the same 1,179 confirmatory answer-phase tutor turns under the frozen helpfulness and pedagogy rubrics. On the primary base under Opus, the policies do not differ significantly in helpfulness but are perfectly rank-separated under the pedagogy rubric (Cliff's $|\delta|{=}0.10$ vs. $1.0$). Across the two judges, pedagogy contrast
arXiv:2607.27547v1 Announce Type: cross Abstract: As generative artificial intelligence becomes one of the most significant systems of knowledge production in our society today, questions relating to who can access and shape that production grow increasingly important in our discourse. This paper argues that the existing frameworks for AI literacy, which are dominated by technical competency and responsible-use principles, are insufficient because they enforce a "consumer" orientation toward AI rather than fostering genuine epistemic agency. Based upon Foucault's concept of power-knowledge, Freire's pedagogy of critical consciousness, and scholarship of digital literacy, this paper proposes a reconceptualization of AI literacy as a critical practice that equips individuals not just to use AI systems, but to critically evaluate them, resist their structuring assumptions, and participate in their governance. The paper further argues that unequal access to AI tools in society recapitulate
arXiv:2607.27232v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly shaping how we consume information and form our worldview. This raises concerns beyond bias in AI: do LLMs grasp the emotional nuances conveyed via textual framing? In this work, we empirically evaluate how well an array of LLMs aligns with human emotional perception. Considering news headlines covering political and geopolitical conflicts, both human participants (n = 3011, a representative sample of the U.K. adult population, via a YouGov survey) and seven LLMs answered whether headlines evoked sympathy for a specified side in a conflict. We find that the correlation between AI and human evaluations varies across models, ranging from very high (0.789, GPT-5.2) to medium (0.4 ,Mistral Large 2512). Crucially, the leading models are broadly aligned with human judgments across all demographic subgroups, including age, gender, level of education, prior geopolitical knowledge, and participants'
arXiv:2607.28550v1 Announce Type: new Abstract: Silicon sampling refers to the use of Large Language Models (LLMs) to generate responses to surveys. It has shown promise, but tends to generate response distributions with unrealistically low variance. We argue that this mode collapse is due to LLMs failure to generate numeric data, and that text responses may be better suited for this task. We analyze whether Semantic Similarity Rating can improve the fidelity of silicon sampling responses when asked about political attitudes. This method solicits text-only responses from LLMs, then maps this to a numeric scale using text embeddings. We find that this method both improves the fidelity of silicon sampling response distributions, and has few parameters to calibrate.
arXiv:2607.28531v1 Announce Type: new Abstract: Football clubs, academies, and federations operate a growing but fragmented portfolio of digital tools: separate systems for video analysis, GPS/performance tracking, medical records, scouting, and administration. This fragmentation is most acute outside the elite European clubs that can afford integration, producing a digital divide that disadvantages grassroots clubs in developing markets such as Brazil, paradoxically the world's largest exporter of professional players. This paper presents, at a conceptual level, the architecture of "AIx4Soccer One Platform," a multi-tenant cloud SaaS operating system that unifies club-management workflows and embeds a structured athlete-development methodology, the PDI Framework (Plano de Desenvolvimento Individual / Individual Development Plan). We describe two companion components: "Tak Tik," a certified two-sided marketplace connecting clubs with video analysts under a 75%/25% (analyst/platform) re
arXiv:2607.28332v1 Announce Type: new Abstract: When AI systems surpass elite human performance and settle into everyday expert practice, the question that follows is how machine judgment is made publicly intelligible and attributable. We study Korean Go commentary on YouTube, where AI systems such as KataGo became standard analytic tools after AlphaGo. Our corpus spans a decade (2016--2025) and approximately $1{,}900$ hours of footage across institutional broadcasters and creator-led channels, in four phases of AI availability. We document a widening asymmetry between visual and verbal AI presence: AI winrate graphs are visible for about $98\%$ of late-period institutional broadcast time, yet AI-salient talk accounts for only $2.63\%$ of sentences. What recedes is the source label, not the metric: winrate and point-gap talk persists while ``AI'' itself goes unsaid. We read this recession as the communicative signature of domestication. Our strongest evidence is a compositional shift i
arXiv:2607.28179v1 Announce Type: new Abstract: This innovative practice full paper examines the integration of technology-enhanced tabletop exercises (TTXs) into computing education, focusing on cybersecurity curricula. The motivation is to better prepare students for complex, collaborative problem solving typical of incident response and IT governance, where coordination, communication, and timely decision-making are essential. Although TTXs are well-established in professional practice, they remain underused in universities. We address this gap by augmenting TTX delivery and evaluation through the INJECT Exercise Platform (IXP), a web-based environment that automates scenario flow and enables data-driven assessment. Our practice implements IXP to automatically deliver scenario updates, facilitate team discussions, and collect interaction data to support automated assessment. This combination enhances realism, reduces instructor workload, and provides actionable insight into student
arXiv:2607.28137v1 Announce Type: new Abstract: Contemporary AI discourse attributes to language models properties they cannot bear: general intelligence as substrate-independent cognition, hallucination as cognitive failure, agency as autonomous goal-pursuit, sentience as emergent inner life, alignment as goal synchronization. This paper argues that these are instances of a single category mistake--properties constituted within human communicative practice are projected onto the machine side--and explains its structure. Human-LLM interaction constitutes a language game in which one side bears all normative activity. We call this configuration asymmetric communication since model outputs circulate communicatively, entering further exchanges, without the system undertaking commitments, bearing entitlements, or performing the assessment on which discursive standing depends. Three conditions define the asymmetry: (i) correctness is enforced exclusively by the receiver; (ii) accountability
arXiv:2607.28041v1 Announce Type: new Abstract: As AI systems become capable of producing the essays, code, reports, summaries, plans, and decisions through which institutions usually recognize competence, a familiar question becomes harder to answer: what is learning for? Existing AI ethics rightly emphasizes present failures--bias, opacity, hallucination, labor extraction, privacy risk, and weak accountability. But if the case for learning rests only on those failures, then each technical improvement appears to weaken it. This article develops a different answer. Using the idealization of AI that executes specified tasks flawlessly while lacking authority over purposes, legitimacy, and responsibility, we argue for post-instrumental learning: learning that preserves the capacities people and institutions need when many useful outputs can be delegated. We analyze five such capacities--end-setting, reason-giving, contestability, refusal/revision, and participation--and name their erosio
arXiv:2607.28023v1 Announce Type: new Abstract: AI accountability at scale is an institutional problem: who can observe, verify, and change deployed systems. We develop a sequential political-economy model in which an AI vendor chooses auditability and substantive mitigation, a deployer monitors after adoption while facing switching costs, and enforcement depends on verifiable evidence. Anticipating the deployer's monitoring response, the vendor may stop at an observable procurement floor while mitigating below the social first best, producing a proxy-compliance equilibrium. We characterize the unique interior equilibrium and the corner in which harm is fully mitigated. Independent audit rights raise enforcement exposure directly; portability restores deployer leverage; incident reporting adds a regulator-visible evidence channel; and outcome-linked liability creates incentives that do not depend on vendor-controlled detection. The results explain why documentation and standardized eva
arXiv:2607.27586v1 Announce Type: new Abstract: As Generative AI (GenAI) tools become increasingly capable of generating solutions to computing assignments, the computing education community is exploring pedagogical approaches that emphasize solution evaluation, verification, and critique alongside traditional solution generation. However, evidence regarding the impact of such evaluation-centered tasks on student learning remains limited, particularly in upper-division, theory-heavy courses. We conducted a randomized A/B crossover study (N=220) in a junior-level algorithms course to compare evaluating GenAI-generated solutions with traditional problem solving. Across six assignments, student working groups either solved challenging algorithmic problems directly or evaluated often-flawed GenAI-generated solutions, with roles reversed midway through the semester. We found no statistically significant differences between groups in midterm scores, final exam scores, overall course grades,
Article URL: https://www.governance.fyi/p/why-singapore-and-estonias-edtech Comments URL: https://news.ycombinator.com/item?id=46825033 Points: 6 # Comments: 3
Article URL: https://github.com/nyr-github/ai-3d-learning Comments URL: https://news.ycombinator.com/item?id=48330451 Points: 2 # Comments: 0
More than a dozen health systems are warning patients of a phishing campaign impersonating Epic’s MyChart portal with fake Medicare rewards and faux login pages. Experts say the scam exposes a deeper problem: patients can’t be expected to spot increasingly polished fakes, so the responsibility falls on providers to communicate in ways that build trust. The post Epic MyChart Phishing Scam Hits Health Systems: What Providers Should Do appeared first on MedCity News .