Papers

Artificial intelligence systems are rapidly becoming critical components in healthcare, finance, public services, and other safety-critical domains. Yet the engineering practices used to evaluate these systems remain predominantly model-centric, emphasizing properties such as accuracy, robustness, fairness, and interpretability before deployment. These properties are necessary but insufficient once an AI system operates within an ever changing socio-technical environment characterized by distribution shifts, institutional constraints, human feedback loops, privacy requirements, and interactions among multiple AI agents. This vision paper introduces AI Deployment Accountability Engineering (ADAE), a proposed AI engineering subdiscipline concerned with establishing measurable, continuous, and actionable accountability for deployed AI systems. ADAE treats accountability as a deployment-layer property rather than solely as a property of an individual model. It seeks to determine whether an AI-enabled system continues to operate within acceptable risk limits, identify the contexts in which failures emerge, attribute failures across interacting technical and human components, translate technical failures into downstream consequences, and support timely intervention. We articulate a research agenda built around four interconnected pillars: structured discovery of context-dependent failure modes, privacy-preserving accountability measurement, system-level risk analysis for agentic AI, and translation of technical failures into operational, and institutional risks. The broader goal is to establish foundational principles, mathematical tools, and system architectures for accountable AI deployment across safety-critical applications.

Teaching abstract theoretical computer science (TCS) concepts such as algorithm analysis and complexity theory is challenging because students must handle formal proofs and asymptotic reasoning that conventional resources rarely explain in an adaptive, on-demand way. We present AlgoRAG, a specialized Retrieval-Augmented Generation (RAG) system that couples a large language model (LLM) with a curated, domain-specific knowledge base to address these challenges. The knowledge base integrates authoritative textbooks, 847 lecture slides, 312 practice problems with solutions, 156 worked proof templates, and 89 complexity worksheets. AlgoRAG incorporates domain-specific optimizations including mathematical entity recognition, notation-aware retrieval, and pedagogical re-ranking. We evaluate AlgoRAG on 179 curated exam-style questions spanning asymptotic analysis, recurrence relations, dynamic programming, graph algorithms, NP-completeness, sorting, and divide-and-conquer. The system achieves a 100% success rate with a mean response time of 38.0 seconds. While BLEU-4 scores are zero – a known limitation of n-gram matching on mathematical proofs where equivalent reasoning may use entirely different notation – AlgoRAG attains ROUGE-1 F1 of 0.0963, ROUGE-L F1 of 0.0683, and a pedagogical quality score of 0.7620, indicating that responses are well-structured and didactically sound even when surface wording diverges from reference answers. Performance is especially strong on NP-completeness (ROUGE-1 F1 = 0.1285, pedagogical quality = 0.7643) and graph algorithms (ROUGE-1 F1 = 0.1023, pedagogical quality = 0.8086). These results support the conclusion that RAG is an effective architecture for personalized theoretical-CS instruction, providing correct, context-rich explanations even for highly abstract topics.

Online attention is commonly summarized using cumulative volume, peak activity, or arithmetic averages, but such measures can obscure differences between activity that is sustained over time, concentrated near the present, or renewed after dormancy. This paper introduces the Time-Horizon Engagement Effectiveness (TH-EE) framework, which constructs interpretable temporal profiles of online attention by distinguishing three related but non-equivalent properties: persistence, recency, and re-emergence. We evaluate the framework through controlled engagement traces, a proof-of-concept application to 1,850 YouTube videos across 37 topics (18 cohorts selected a priori as exemplars of persistent, acute, cyclical, and recently originating attention, and 19 cohorts corresponding to authoritatively debunked claims), and an event-level validation on eleven years of daily Wikipedia pageview series for the same topics. The controlled analyses show that the framework distinguishes distributed activity from concentrated bursts, introduces temporal-order sensitivity through recency weighting, and identifies renewed activity after a defined dormant interval. On the event-level series, the framework's reactivations co-locate with Kleinberg burst onsets, and PELT change points far more often than chance. The YouTube application shows that debunked-claim cohorts do not occupy a unique region of temporal-profile space; they exhibit heterogeneous patterns that overlap substantially with benign topics. These results support treating persistence, recency, and re-emergence as separate dimensions of online attention. TH-EE is a descriptive and comparative measurement framework, not a classifier of misinformation, coordination, intent, or content veracity.

When the same question is asked of multiple AI advisers, as in self-consistency and LLM-as-a-judge panels, Condorcet's jury theorem predicts that adding independent, competent advisers makes the majority more reliable. The theorem, however, has a latent dimension when viewed from the user's vantage: adding advisers also makes disagreement more visible. A binomial model reveals that this “visible dissent” becomes nearly inevitable as the number of advisers grows, and that reliability and disagreement approach certainty at rates that cross at an adviser accuracy of 4/5 (0.8); below it, visible dissent eventually becomes more likely than a correct majority. Even ideal panels of independent and competent advisers can be correct in aggregate but appear divided; such disagreement does not by itself indicate aggregation failure. The way advisers split also provides a common basis for predictive multiplicity, reconciliation load, and reliance miscalibration. These results separate aggregation from disclosure and turn the latter into testable questions about how disagreement should be presented and interpreted.

Predicting a user's next POI is a task in human mobility modeling, yet LLM-based approaches focus on semantic reasoning from previous mobility records, while neglecting real-world spatial context. However, human mobility is inherently shaped by spatial cognition, including geographic distance and neighborhood context. This issue is further compounded by prior evidence that LLMs often struggle with spatial reasoning tasks, including distance estimation and geographically biased prediction. To address these limitations, we propose our framework, a multi-agent LLM framework that decomposes next-POI prediction into three stages: Firstly, a Pattern Extraction Agent that captures temporal and categorical mobility patterns from trajectory history; Secondly, a Spatial Reasoning Agent that structures candidate activity choices by combining behavioral preferences with real-world spatial constraints, including geographic distance, road network distance, and neighborhood affiliation; and Thirdly, a Decision Synthesis Agent that integrates behavioral patterns and spatial reasoning for final prediction. Experiments on the NYC benchmark dataset with two LLM backbones show improvements over baseline methods, with up to 493% Hit@1 improvement and 37% relative improvement in Hit@5. Ablations show that combining neighborhood affiliation with distance-based features generally outperforms distance-only settings, and that the Spatial Reasoning Agent plays a crucial role in final prediction by integrating behavioral preferences with real-world spatial constraints, especially for smaller models. Overall, the results highlight the importance of spatial reasoning in mobility prediction. Accurate next-POI prediction requires combining behavioral patterns with explicit real-world spatial constraints, and multi-agent decomposition provides an effective structure for organizing these forms of context.

Low-cost indoor air-quality sensors could support personalized asthma prevention, but their nonlinear measurement error, delayed exposure effects, time-varying confounding, and heterogeneous lower-tail responses limit risk estimation. We present CAUSALQUANT-ASTHMA, a measurement-error-aware causal quantile distributed-lag framework for short-horizon peak expiratory flow analysis. Sparse reference measurements train a nonlinear calibration model; stabilized sequential generalized-propensity weights address measured exposure assignment; and a susceptibility-modulated, smooth, noncrossing quantile model estimates lag-specific and sustained-exposure contrasts. Because no authorized cohort simultaneously provided dense indoor sensing, reference co-location, and outcome-compatible longitudinal data, evaluation used five semi-synthetic panels with known counterfactual truth, 150 patients and 12,600 patient-days per realization. Across eight methods, CAUSALQUANT achieved a dose-response integrated absolute error of 0.304 plus or minus 0.094, improving 24.2 percent over the strongest measurement-error and propensity-weighted baseline. It also obtained the lowest overall pinball loss, 1.065, while maintaining zero quantile crossings and 78.1 percent coverage for the nominal 80 percent interval. Sensor calibration reduced held-out exposure RMSE by 33.7 percent. Stress tests quantified degradation under sensor noise, missing personal measurements, and hidden confounding. These findings establish methodological feasibility and reproducibility, not clinical effectiveness; prospective, governance-approved external validation is required before patient-level interpretation or deployment.

Artificial intelligence helps education most where an essential provision has been rationed by cost. For language learners that provision is a teacher's voice, which binds listening, reading, speaking, and writing into one act. Published evidence shows why most learners lack it, from a global shortage of 44 million teachers to heavy household tutoring bills, and why technology has not substituted for it: computer-assisted language learning proved effective but narrow, applications presuppose connectivity 2.6 billion people lack, and One Laptop per Child's randomized evaluation found that hardware without capable software teaches nothing. We distill eight difficulties and four binding constraints, and argue that small open-weight models dissolve the last: a complete four-skill stack now fits a \$200-class laptop and, on community measurements, generates at the pace speech is consumed, for about one US cent of electricity per study hour. We therefore propose LLMersion, a scheme for AI for education that runs entirely at home, over the learner's own documents, with an AI-written, AI-understood, AI-updated codebase anyone can customize; present LLMersion-1, a released open-source prototype (https://github.com/QM378/LLMersion ); and outline the vision of a private learning agent.

Visual impairment affects over 2.2 billion people worldwide, yet conventional white canes cannot detect elevated hazards or provide semantic environmental context. Existing AI-assisted navigation systems typically rely on expensive hardware or cloud connectivity, limiting accessibility in resource-constrained settings. This paper presents an affordable (\$88 USD), fully offline AI-integrated smart cane designed for multimodal mobility assistance on an ultra-low-power Raspberry Pi Zero 2W. The system fuses RGB vision sensing with Time-of-Flight (ToF) distance estimation, pairing an INT8-quantized SSD MobileNet V1 model with distance-aware vibrotactile feedback and real-time audio alerts. To ensure operational robustness on constrained hardware, a multiprocessing architecture isolates sensor acquisition, neural inference, and haptic feedback into independent processes with fail-safe sensing support. Experimental evaluation across indoor mobility scenarios demonstrates a macro-averaged F1-score of 0.82 (precision: 0.85, recall: 0.81), a mean end-to-end latency of 330\,ms, and a peak power draw of 2.8\,W. A preliminary usability study with 12 participants (SUS: 78.5, NASA-TLX) demonstrated positive user perception and enhanced obstacle awareness. The proposed prototype validates the feasibility of deploying privacy-preserving, edge-native assistive intelligence for cost-sensitive mobility assistance.

The rapid diffusion of hate speech and misinformation on social networks challenges democratic societies, since direct suppression efforts may deepen polarization, fuel public distrusts, and strengthen extremist narratives. LLM-driven counter-narratives (CNs) offer a promising way to reduce those risks, yet their effectiveness depends on rhetorical and stylistic choices that remain poorly understood. We present a multi-stage agent-based framework for generating, refining, and evaluating CNs, applied to pro-Russian hate and misinformation narratives on the war with Ukraine and adaptable to other domains. A pilot experiment with human evaluators identifies effective technique style pairings, such as repetition with emotional framing enhancing persuasiveness. Building on these insights, we introduce a multi-agent refinement process that iteratively improves CNs for persuasiveness, emotional engagement, and shareability. After human validation confirmed improvement, an automated safety analysis shows that our refined CNs match or improve on expert-written counterspeech. A simulated experiment then shows that they reduce the perceived strength of pro-Russian narratives and consistently outperform a vanilla LLM baseline, highlighting a pathway toward scalable, narrative-specific interventions against hate speech and misinformation. Code and data accompanying this work are publicly available at https://github.com/carmelkron/inlg2026-counter-narratives.

Social networks shape how individuals make decisions and how opportunities are distributed. However, the mechanisms that generate these networks often reflect pre-existing inequalities, and technologies that rely on network-derived signals risk further amplifying such disparities. Algorithmic fairness research largely treats networks as a fixed background, grounding analysis almost exclusively in distributive justice and overlooking how network structures systematically bias decision-making. In this Perspective, we identify ten network effects and trace how they create structural biases in the relationship between what we intend to measure and what we observe. Using academic hiring as an example, we show that network biases are not inherently harmful or beneficial. Determining their legitimacy requires examining the entire decision-making process through the lenses of both distributive and procedural justice while engaging all affected stakeholders. We therefore call for a holistic, networked approach to fairness that moves beyond static group categories and recognizes the dynamic, relational, and structural nature of inequality.

Large Language Models (LLMs) are powerful zero-shot learners but remain prone to misalignment with human preferences, often producing biased, toxic, or otherwise harmful outputs. Existing alignment methods, while effective, are costly and tightly coupled to the model, limiting flexibility and scalability. We propose a modular correction framework that augments pretrained LLMs with Activated LoRA (aLoRA) adapters and a context-aware routing mechanism to eliminate harms from misaligned model responses. Our approach enables expert adapters to activate mid-sequence without invalidating the KV cache, allowing low-latency, targeted correction during generation. Each expert is trained to detect and mitigate specific harms, such as bias or toxicity. A learned router dynamically selects appropriate experts based on the models intermediate outputs. We demonstrate that our system improves alignment on standard safety benchmarks while preserving task performance, offering a lightweight and efficient path toward safer and more controllable LLM deployments.

Age verification is often treated as a technical problem: can a system determine a child's age accurately? We argue this framing is too narrow. Age assurance becomes consequential when evidence is evaluated, translated into age-related claims, and used to decide whether a person can access, purchase, or belong. We review 85 publications on children's age assurance published from 2020 through February 2026. We find that shared terms such as age verification describe different processes. Age is represented as threshold eligibility or inferred estimation, and the same eligibility claim can arise from different evidence and components. Rights, access, and privacy receive more attention than accuracy, error, and fairness; yet institutional actors are rarely connected to system failures or user remedies, an accountability gap. We introduce the Age-Assurance Process Framework, which treats age assurance as a sociotechnical process connecting evidence, evaluation, claims, and decisions rather than reducing it to a technical problem.

Safety research often focuses on model-generated harms, but users may also direct hostility, coercion, and adversarial pressure at models. Understanding how and when that occurs is essential for accurately interpreting model behaviour, alignment drift, and real-world deployment risks. In this paper, we audit 777K English LMSYS-Chat-1M conversations with two independent detectors: an eight-category lexicon for hostility directed at the model, and the dataset's moderation signal; and show that they capture different, weakly overlapping phenomena. The lexicon identifies insults, threats, and jailbreak coercion aimed at the assistant, while moderation flags are dominated by toxic-content solicitation rather than hostility at the model. Together, they mark about 5% of user turns; adjusting the narrower lexicon-harassment union for measured precision puts mistreatment aimed at the assistant at 0.90%. These absolute rates describe arena-style evaluation traffic and should not be read as deployment-wide base rates. We find that user hostility varies 13-fold across models, driven largely by who each model attracts rather than by model behaviour: first-turn hostility spreads far wider than post-response hostility, and more than fifteenfold separates the extremes even after deduplicating opening prompts. Within conversations, assistant apologies are consistently associated with higher odds of next-turn hostility under both detectors; the effect survives restricting to non-refused prior turns and to jailbreak-free conversations, and is positive in 20 of 23 models. Yet across models, more apologetic models receive less hostility overall. Finally, hostility also shows temporal structure, with coercive openings front-loading the first turn while affective hostility accumulates over a session. We release the lexicon, the detector cross-validation pipeline, and all derived tables.

AI chatbots are increasingly used as sources of emotional support, on dedicated companion apps and general-purpose assistants alike, yet little is known about what happens when users try to leave. Combining a content analysis of Reddit posts about quitting or reducing use (N=2,782) with interviews with users who found leaving difficult (N=16), we show that disengagement sometimes is not a single decision but a recursive trajectory: triggers prompt users to question the relationship, attempts to leave collide with barriers, and some users cycle through quitting and returning. We propose the notion of the addictive intimacy of AI, a configuration in which the qualities that make a companion emotionally valuable are the same ones that make it harder for users to limit their use and leave, so that intimacy and disengagement risk cannot be treated as independent design problems. We close with design implications for responsible offboarding.

K-12 students increasingly form relationships with AI companions. Schools face growing expectations to teach AI literacy, yet existing frameworks treat AI as a tool rather than a relationship, and little is known about how teachers understand and act on students' relational use of AI. We conducted scenario-based interviews with 33 US K-12 teachers. Teachers welcomed academic companions but worried that intimate companions remove the developmental friction through which students learn to sustain human relationships. Teachers drew the boundaries of their jurisdiction by setting and observable wellbeing: within it they taught, talked, and watched; beyond it they positioned themselves as the adults best placed to notice and connect students with support. They envisioned AI companion literacy as shared work across the jurisdictions of counselors, parents, platforms, and policymakers, spiraling across grade levels. We introduce AI companion literacy as an extension of AI literacy and discuss implications for K-12 AI education.

This volume contains the accepted lightning-talk contributions from the 2nd International Workshop on Low Carbon Computing (LOCO 2026), held at Lancaster University, United Kingdom, on 10-11 September 2026. The collection brings together 14 short papers presenting emerging research, early-stage results, practical tools, and critical perspectives on reducing the environmental impacts of computing. Topics include carbon and energy measurement, carbon-aware scheduling, sustainable artificial intelligence and machine learning, high-performance computing, computational sufficiency, hardware longevity, interoperability, frugal computing, and post-growth approaches to digital technology. The contributions were selected by the LOCO 2026 Programme Committee for their relevance to the workshop and their potential to stimulate discussion across technical, organisational, and societal dimensions of low-carbon computing. They are published together in this edited volume as a companion collection to the main LOCO 2026 proceedings index.

Robots designed to mediate human groups often fall into the solutionist trap: they are framed as sociable agents that fix problems such as conflict, disengagement, or lack of coordination. We suggest a different way of thinking. Rather than discrete agents, robots can be understood as situated elements of shared environments; catalysts and carriers of group experience whose meaning emerges through how people position, interpret, and interact with them. From this perspective, robots are not there to repair some ostensible dysfunctionality, but to enable group-level sense-making around care, norms, and identity. Our prior work on robotic street furniture suggests that this does not happen by imitating human sociality but by taking the shape of deliberately constrained, group-facing entities that happen and act for us without being socially entangled as one of us. We thus understand robots in public spaces not in terms of autonomy or intelligence, but as a relational capacity. This implies designing robots not in our image or for our utility, but grounded in our needs in being and becoming together.

Proposed governance framework for autonomous robotic systems, introducing a three-layer compliance architecture (ARC) instantiated through model safety validation, cognitive certification benchmarks, and operational authorization standards.

Global estimates of access to health care treat every facility alike. Drive time was routed on the OpenStreetMap network from all 217,841 of its hospitals and 15-, 30- and 60-minute isochrones dissolved for nested capability tiers, weighted with GHS-POP 2025. 96.2% of the world's population lives within an hour's drive of a hospital, 78.5% of a hospital with a documented emergency department, and 40.3% of one with a helipad or university name marking tertiary care. The 1.45 billion people within an hour of a hospital but not of an emergency hospital live mainly in South and East Asia and Nigeria; the 312 million beyond an hour of any hospital lie in a belt from the Sahel to Madagascar. Defining the tertiary tier by helipad alone leaves its estimate unchanged, and the isochrone construction shifts the 60-minute figures by a few percent. Counting every facility as equivalent overstates access to time-critical care.

Freelance game coaches monetize their gaming expertise by offering personalized instruction to players seeking to improve, working through gig platforms, yet little is known about how they operate. To address this gap, we conducted semi-structured interviews with 20 experienced freelance coaches across 17 competitive games on Fiverr. Despite lacking shared formal training, these coaches converged on similar practices centered on rapport-building, individualized diagnosis, and adaptive feedback. We identify two structural conditions shaping this work: dual precarity, in which coaches navigate both gig platform instability and the lifecycle volatility of live-service games; and earned authority, in which coaches must continually establish legitimacy through visible competitive achievement within the same gaming spaces as their students. These coaches welcomed AI for administrative and analytic support but resisted its use in live interactions where trust, relational engagement, and situated judgment remained central. We discuss implications for Games HCI and the design of computational coaching systems.

Mindspeller produces a Neuroprofile from three sources: rational self-report, association-based semantic positioning, and performance recorded during cognitive tasks together with EEG. The task-and-EEG stream is the only source used to create occupational evidence. A result can enter role matching only after the participant's task performance supports the intended construct and the corresponding EEG data pass the required quality and evidence checks. The current pilot uses four EEG electrodes, twelve scored tasks, 23 O*NET abilities, and an internal bank of 376 occupations. Self-report and association evidence help explain motivation, preference, and alignment, but do not generate roles. The output is intended to support discussion about cognitive fit. It is not a hiring decision, a measure of practical job skill, or a prediction of job performance. Role confidence is currently capped at Moderate, and external psychometric and job-outcome validity have not yet been established.

We aim to characterise the value of artificial intelligence in the workplace. Current studies largely measure this value in terms of the current automation capabilities and public adoption of AI. However, such metrics ignore the greater impacts of human–agent collaboration in transforming the nature of work. To account for this, we must expand the scope of our analysis beyond atomised tasks of today, and instead focus on how AI can augment entire workflows of the future. To ground this analysis, we establish a precise definition of AI augmentation comprising six conditions, spanning durable net value, meaningful human control, accountability and recovery, and long-term human development through learning, career pathways, and job purpose. We elaborate on these conditions and apply the framework in a case study of AI-mediated social surveys. We conclude by outlining how organisations, researchers, and government leaders can use this framework to make sense of the future of work.

Did people stop asking other people for advice online once generative AI could answer their questions? Prior work on ChatGPT's effect on online help-seeking disagrees in both size and sign, in part because no study has compared affected communities against similar communities that AI cannot easily substitute for, over the same months. In this paper, we track monthly post counts in 26 Reddit informational communities against 90 size-comparable hobby communities over the same six calendar months before and after the launch of ChatGPT. We also repeat the entire analysis at 66 earlier dates, before ChatGPT existed, to see what our method reports when no ChatGPT-effect exists. We find that informational help-seeking did not decline. Our results rule out any decline in posting larger than 3.4%, far smaller than the 8% to 25% declines documented in prior work. Steady post counts could still be misleading if AI-written posts had replaced human ones. We test this possibility by scoring 274,411 posts and 223,775 comments with AI-text detectors, compared in a way that cancels out detector false-positives on human-written text. AI-written posts rose only 2-3 percentage points more in informational communities than in hobby communities, short of the 5.1 points that would be needed to hide even the smallest decline previously reported for Reddit. In addition, the comments people receive show no such rise at all. Why, then, do published studies disagree? Reddit community types were already drifting apart before ChatGPT existed, at rates comparable to every published estimate, and without same-time controls, that drift can look like an effect of generative AI. Our own largest estimate, an 18% fall in posts to low-stakes curiosity communities, matches its pre-existing trend. Humans still ask humans for help, and, as far as detection can tell, humans still answer them.

Large language models (LLMs) are proposed as synthetic respondents for pilot testing, but their usefulness depends on whether they supply the evidence assessment development requires. We calibrated rating scale models on 14 digital-use skill items from 6,245 adults and used the human item parameters to evaluate responses generated for 1,300 demographically matched personas. LLM responses had high internal consistency ($\alpha \approx .94$) but used the lowest category in 0.2-0.3% of responses versus 13.7-21.5% for humans, and no persona chose it on every item. These gaps changed the response-category judgment on both subscales and the PC targeting judgment; on the human-calibrated scales, 12 of 14 items had infit below 0.70, indicating responses more predictable than the Rasch model expects. Preregistered changes to the prompt, category order, and persona information did not restore the human lower range. High internal consistency is insufficient evidence that LLM responses can replace human pilot data.

Generative artificial intelligence (AI) is increasingly used to support human decision-making, yet less attention has been paid to how AI may influence the reasoning processes that precede final judgments. This pilot study explored whether different AI-use conditions were associated with differences in reasoning breadth during ethical decision-making. Twenty-nine participants completed an ethical dilemma under one of three conditions: AI-Prohibited (n = 10), AI-Optional (n = 10), or AI-Mandatory (n = 9). Responses were evaluated by three independent blind coders using two exploratory measures: the Counterargument Diversity Score (CDS) and Perspective Diversity Index (PDI). Participants across conditions generally converged on similar ethical conclusions, most commonly favoring disclosure and customer protection. However, participants in the AI-Mandatory condition considered a broader range of perspectives, including legal, regulatory, organizational, technical, and ethical viewpoints. A statistically significant overall difference in PDI scores was observed across the three conditions, whereas differences in CDS were not statistically significant. These findings suggest that generative AI may not necessarily alter final ethical judgments but may be associated with broader exploration of perspectives prior to reaching those judgments. Given the small sample size, the findings should be interpreted cautiously and examined in larger studies.

Electronic health record (EHR) foundation models trained on longitudinal patient trajectories have demonstrated strong performance across diverse clinical prediction tasks. However, their clinical reasoning capabilities remain constrained by next-token prediction on limited and incomplete EHR data. To address this, we propose a reinforcement learning (RL) fine-tuning framework that treats EHR foundation models as generative policies over patient trajectories. We formulate common clinical prediction problems (e.g., hospital readmission) as event-conditioned, time-windowed reasoning tasks. We then design time-aware, rollout-sensitive rewards to account for finite rollout lengths and temporally inconclusive outcomes. We find that RL fine-tuning consistently improves over pre-trained backbones and strong baselines. Notably, it enables smaller models to surpass larger pre-trained models in data-limited regimes and induces positive transfer across tasks. Further analysis shows that RL fine-tuned models generate trajectories with stronger structural and semantic alignment to ground truth and greater downstream utility.

Braille literacy among blind school-age children has fallen sharply, in part because producing illustrated braille pages still requires specialized software and trained labor. We present Tact, a browser-based pipeline that converts a spoken or typed story idea into printable braille with a matching raised tactile illustration, without an account or mandatory cost and with an offline-capable path. The paper documents the engineering history of the system: its sighted-operator ethical model; hardware rationale for consumer fused-deposition modeling; physical braille geometry and printer calibration; local, hosted, and fallback language-model paths; a deterministic Grade 1 braille translator; verified page layout and pagination; a 93-shape hand-drawn tactile illustration library; and synthesized sound design for a voice-first interface. We report engineering verification, ethical commitments, limitations, and the work required before the system is ready for real blind and low-vision readers.

Super-apps, an emerging mobile architecture, host third-party mini-apps inside a single app, allowing users to access diverse services. A decade of security research on the super-app ecosystem has all assumed super-apps to be a trusted intermediary. We argue this implicit trust is difficult to justify: China's WeChat is already shown to passively track its user's activity across mini-apps at extraordinary scale; Russia's MAX's parent company is reported to be deeply entangled with the state prosecution of online speech; and Iran's Bale was reported to be functioning in the world's longest internet shutdown due to its state-backed support. In this paper, we show how malicious super-apps have undeniable capabilities to silently undermine the security and privacy of mini-apps and users without leaving any trace. Using MAX as an example, we show how it can capture mini-app UI, read and write mini-app local storage, inject arbitrary JavaScript into a mini-app's runtime, mediate mini-app network traffic, and control authentication context in ways that can enable silent user impersonation. Sadly, these capabilities manifest themselves in any super-app because of the architectural privileges granted to them by design. We argue that mobile OS and app store interventions are urgently needed to close this architectural blind spot before it is further exploited.

Reproducible research practices are context engineering for AI coding agents. I argue that agents lower the cost of maintaining tests, commit histories, repository structure, instructions, and decision records while making their benefits immediate. Researchers remain responsible for verifying these artifacts and the scientific judgments they encode.

Generative artificial intelligence is entering transportation through traveler-facing advisories, synthetic crash-record generation, and policy decision support. Existing governance frameworks lack transport-specific statistical tools to measure distributional risks across heterogeneous populations. We develop a Distributional Sociotechnical Audit (DSA) that integrates algorithmic equity, synthetic-data validity, and public-attitude heterogeneity into one empirical pipeline. The audit analyzes 5,760 persona-controlled queries to four LLM families across 12 demographic cues and four transport topics, uses two cross-family judges and a Wasserstein-2 Equity Dispersion Index, tests three FARS crash-record generators with conditional projected maximum mean discrepancy (cpMMD), fits a Bayesian ordered-logit model to Pew American Trends Panel Wave 152 (N = 4,538), and combines the signals into a continuous Sociotechnical Risk Index. Congestion-pricing advice has the highest persona-based dispersion (mean EDI = 1.96; highest direct EDI = 2.20). CART synthetic crash records fail all conditional tests (p < 0.001), while the Gaussian copula has borderline conditional stress (p = 0.105) despite passing marginal checks. Attitudes to AI vary across demographic strata. Distributional audits and continuous risk indices with sensitivity reporting offer a more defensible basis for transport GenAI governance than categorical approval tiers, which show a 75% assignment flip rate under weight perturbation.