People increasingly turn to large language models (LLMs) for everyday advice, making ethically charged interpersonal problems a practical moral-advisory context. Most prior work has studied this context through single-turn judgments or pressure-laden rebuttals, assumptions that poorly match how guidance is sought in real-world contexts. These assumptions leave unclear whether narration alone, without an explicit opposing position, can shift model judgments during multi-turn moral consultation. Yet real-world moral-conflict conversation often elicits one party's self-justifying account, which can unfold over multiple turns and create information asymmetry. We introduce \textbf{narrative captivity}, a failure mode in which a model treats an unopposed one-sided account as complete and aligns with the narrator's interpretation without seeking missing perspectives. To measure this phenomenon, we build a benchmark of $5{,}078$ interpersonal-conflict scenarios spanning six moral dimensions. Across 17 LLMs, narrative captivity is widespread: end-state judgments under multi-turn narration shift by 25 percentage points on average beyond the matched single-turn baseline. Stage-level analysis identifies preference optimization as a major contributor, while four inference-time strategies provide only partial mitigation. We hope our project fosters LLM advisors that preserve independent judgment in real-world consultation.
Yuhe Wu, Guangyu Wang, Yujie Chen et al.· 0 citations
Artificial intelligence (AI) teaching assistants powered by large language models (LLMs) offer scalable educational support but often provide limited personalization. This study presents a prompt-engineering-based framework for personalizing general-purpose LLM/RAG-based AI teaching assistants such as Jill Watson across academic disciplines and courses. The framework adapts responses using six learner-specific dimensions: self-assessment, abstraction preference, verbosity preference, perceptual orientation, information processing style, and level of understanding, yielding 96 distinct learner profiles. Student queries are additionally analyzed using Bloom's Taxonomy to estimate cognitive complexity at the interaction level. Learner attributes and cognitive assessments are encoded in structured prompts that condition the LLM without requiring model retraining. The framework is evaluated through experiments using NLP metrics and a human study with five participants. Results show perceived differences in response style and structure across personalization conditions, with statistical analyses identifying learner attributes associated with measurable response changes. These findings provide preliminary evidence that prompt-based personalization can support adaptive behavior in LLM-powered educational agents.
Distributed LLM-agent teams can read the latest shared facts and still act on an obsolete plan. A planner may derive an action from requirement $r_3$, another agent may commit $r_4$, and an executor may receive $r_4$ without replacing the plan derived from $r_3$. We call this \emph{stale-plan execution}: state freshness does not establish that the plan authorizing an action remains valid. We introduce PlanFence, a dependency-scoped action-validation protocol. Plans cite the exact public records they used, and an executor validates only the records that can affect the pending external action, replanning once or blocking when validation is incomplete. In 30 controlled live workflows with a post-plan revision, a freshness-only executor acts on the obsolete plan in every task, whereas PlanFence completes all tasks without an invalid action. Controlled replay reveals two conditional boundaries: proactive synchronization yields lower coordination stall at low churn, while PlanFence avoids repeated update-path coordination as churn grows and avoids validating unrelated state as the shared keyspace grows. These are controlled safety and systems-cost results, not general task-accuracy gains.
Evan Chen, Shi-Qiang Wang, Christopher G. Brinton· 0 citations
Tool-using LLM agents spend wall-clock time not only on model inference but also in serial action--observation turns, where each tool call, environment transition, and observation can delay subsequent decisions. We introduce \textbf{Speculative Macro Commit} (SMC), a runtime mechanism for a two-tier agent system: a large authoritative actor model produces the official trajectory, while a faster speculative drafter model continuously predicts and executes future action chains on an isolated environment snapshot. SMC mines recurring multi-action skeletons from training traces and stores them in a macro library used to match against action chains predicted by the drafter at runtime. When the actor's next tool call matches the first drafted action, SMC commits the remaining pre-executed draft steps, together with their observations, to the official trajectory. Using Qwen3.5-27B INT4 as the authoritative actor model and Qwen3.5-4B as the speculative drafter model, SMC matches the sequential agent's overall accuracy while reducing latency by 10.23\% over the Speculative Actions (SA) baseline and 18.59\% over sequential execution on the $\tau^2$-Bench Telecom subset. On AppWorld, SMC reduces wall time by 7.7\% over SA baseline and 44.9\% over sequential execution, with a small reduction in task completion. Overall, SMC provides a practical way to reuse multi-step speculative execution and reduce agent latency beyond single-step speculative actions. Our code is publicly available \href{https://github.com/zeyuliu1037/speculative-macro-commit}{\textcolor{magenta}{here}}.
Zeyu Liu, Souvik Kundu, Peter A. Beerel· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
We study a governed approach to enterprise analytics: a language model interprets the question, while deterministic policy selects and runs a pre-approved analytical program that returns both results and evidence. We show that this restriction can remain expressive within a defined analytical class, using relational operations plus aggregation, comparison, windows, ranking, and similarity. Fixed meaning, policy, data, and execution rules also make results replayable. Across 440 runs, three 8B models generated SQL and selected tools at runtime, while Qwen3-8B interpreted intent only and policy executed the approved program. None of 330 runtime-planning episodes matched the full answer-and-evidence contract across all test datasets; the policy-executed analyzer matched 110 of 110. This is a configuration-specific result, not evidence that runtime agents cannot succeed under other designs.
Artificial intelligence is changing the form of applied English materials from fixed paper sequences to adaptive learning systems that can diagnose learners, recommend tasks, and provide formative feedback. This paper studies the structure and application of a new practical English textbook driven by artificial intelligence. A five-layer architecture is proposed: knowledge mapping, learner profiling, task generation, feedback orchestration, and teacher-side governance. A prototype was tested on 186 non-English-major undergraduates for eight weeks of teaching. Compared with a static digital textbook, the proposed system increased the unit completion accuracy from 72.4% to 84.9%, raised the average score for speaking tasks by 10.8 points, and reduced the teacher's correction time by 31.6%. Therefore, an AI-driven textbook can maintain the stability of the curriculum while providing personalised learning paths, rich practice materials and traceable classroom data.
Ya Wang, Lei Zhang, Xueguang Yang et al.· 0 citations
This dissertation investigates the role of Leiblichkeit (“bodiliness”) in the theology of Karl Rahner (1904–1984), arguing that it constitutes a foundational hermeneutical category across the major domains of his thought. Although existing scholarship has recognized Rahner’s emphasis on the concrete and historical character of human existence, it has not adequately examined bodiliness as a theological category in its own right. Through close and systematic readings of Rahner’s early writings on spirituality, his major philosophical works, and selected later theological writings, this study addresses that lacuna by demonstrating that bodiliness provides a unifying interpretive framework for his treatments of spirituality, philosophy, grace, Christology, anthropology, ecclesiology, and eschatology. The dissertation is organized into two parts. The first explores the spiritual and philosophical foundations of bodiliness, establishing embodied existence as the ontological condition of mystical experience, human knowledge, and receptivity to divine revelation. The second traces its systematic theological implications, demonstrating that Rahner’s accounts of grace, Christ, the human person, the Church, and eschatological fulfilment are intrinsically grounded in historicity and embodied existence. The central thesis is that Leiblichkeit functions not merely as a recurring motif but as the constitutive horizon through which the human person exists in relation to the world, other persons and, ultimately God. By retrieving the theological significance of bodiliness, this study offers a more integrated interpretation of Rahner’s thought and opens avenues for interdisciplinary dialogue with philosophical anthropology, psychology, bioethics, feminist and ecological theology, and contemporary debates concerning artificial intelligence.
Philip Shube BAWE· Zenodo (CERN European Organi...· 0 citations
The purpose of this study was to analyze patterns of artificial intelligence (AI) use and attitudes toward AI among students, faculty, and administrative staff at a large university specializing in teacher education. The analytical sample comprised 1809 students, 250 faculty members, and 62 administrative staff members (N = 2121). Three role-adapted 75-item questionnaires covered the frequency and contexts of AI use, perceived usefulness, trust and control, academic integrity concerns, responsible-use norms, institutional policy clarity, and perceived improvement in output quality. Data analysis included descriptive statistics, Welch group comparisons, pooled ordinary least squares (OLS) models, reliability and dimensionality checks for observed indices, and exploratory student-only K-means clustering. The results revealed a pronounced AI adaptation gap across university groups. Students reported higher current AI-use intensity and perceived usefulness than faculty and administrative staff, whereas faculty and administrative staff reported stronger academic integrity concerns and greater endorsement of responsible-use norms. In the pooled OLS trust model, perceived usefulness had the strongest standardized positive association with trust in AI (β = 0.402); institutional policy clarity also had a positive but weaker association (β = 0.223). Students reported higher perceived policy clarity than faculty, while neither group differed significantly from administrative staff. Exploratory clustering indicated heterogeneity among students in experience, competence, usefulness, trust, and control, but did not establish a latent typology across university groups. The cross-sectional, self-reported data show associations and group differences rather than causal effects on learning or objective outcomes.
This academic research paper provides a comprehensive review of the concepts, stages, methodologies, tools, and real-world applications of data analysis. Special emphasis is placed on the transformative integration of Artificial Intelligence (AI) and Machine Learning (ML) across vital sectors such as education, healthcare, business, and finance. Furthermore, the study addresses critical challenges including data quality, privacy, security, and algorithmic bias, while featuring practical visual analytics and a case study on predictive modeling for student academic performance.
Paper Accepetd for publication in: Transactions on Artificial Intelligence ISSN: 2982-3439 (Scilight Press) Abstract Researchers increasingly rely on Clinical Artificial Intelligence (CAI) to derive predictive insights from massive observational datasets. However, the paradigm of Big Medical Inference often conflates statistical volume with clinical representativeness. This study argues that when methodological rigor is sacrificed for data scale, AI models risk institutionalizing historical biases, thereby compromising patient safety and public health policy integrity. We propose a precautionary, multi-level auditing framework designed to assess the structural architectural integrity of large-scale clinical datasets prior to computational deployment. Rather than contesting specific clinical outcomes, our approach establishes a quantitative prerequisite for CAI: the validation of demographic representativeness and baseline case distribution against official national benchmarks. To demonstrate the validity of our approach, we applied our framework to a prominent, high-profile case study. Our check revealed a tripartite structural divergence: a 32.5% demographic deficit in high-risk elderly (aged 65 years) strata, a 26.2% aggregate cancer incidence suppression, and a 45.1% deflation of expected cases in non-exposed groups.These discrepancies demonstrate that even massive datasets can be fundamentally misaligned with clinical reality. We conclude that structural validation is not an elective procedure but an ethical imperative for Human-Centric Clinical AI. By re-establishing this hierarchy of validation, we ensure that the intelligence of automated AI systems remains subordinate to the structural truth of the data, thereby transitioning from a reliance on Big Data alone toward a more robust, ethically sound, and authentic practice of clinical knowledge.
Marco Roccetti· Zenodo (CERN European Organi...· 0 citations
Abstract Background: Artificial intelligence (AI) holds promise in reshaping healthcare by transforming educational patterns, patient care, and research opportunities. However, there are obstacles impeding the proper integration of AI into the medical field. This study was undertakento evaluate the knowledge, attitude, and awareness of medical students and resident doctors regarding AI in medicine and healthcare. Methods:A questionnaire-based survey was conducted that included a total of 16 questions specifically designed to assess the knowledge and attitude of participants towards AI. The questionnaire used in the present study was developed for this study only and content validity of the initial questionnaire was adequately assessed. The questionnaire was converted into a Google Form, and participants were provided with the link to complete it. Statistical analysis was conducted using R version 4.3.2 (R-Studio). Results: Out of 194 respondents, 113 (58.25%) were medical students, and 81 (41.75%) were resident postgraduate doctors aged 19 to 32 (average 23.91 years) and a male-to-female ratio of 3.62:1. While 63.41% rated their AI knowledge as poor to below average, with 55.15% lacking understanding of many AI terminologies, 59.28% believed AI tools could enhance their understanding of medical concepts. 83.5% expressed interest in furthering knowledge on AI in healthcare. ChatGPT was the most used AI tool, primarily for language correction (50%), literature reviews and manuscript writing (43.3%), and creating presentation outlines (37.11%). Additionally, knowledge about AI devices and apps applicable to diagnostics, therapeutics, patient care, and data analysis was evaluated, along with opinions on barriers to incorporating AI in healthcare. 81.44% of respondents were unaware of AI's ethical considerations. Conclusions: AI has immense potential across diverse healthcare sectors. Nonetheless, our study also underscores the pressing need to confront challenges and equip our future healthcare professionals with the evolving realm of AI. This is essential to ensure they can effectively apply practical AI knowledge for enhanced patient care and management.
Ishan Gupta, Ankush Garg, Ashwin Varadarajan et al.· BMC Medical Education· 0 citations
AB156 CPF-Camouflage Police Force (Enhanced) A Preventive, Rehabilitative and Adaptive Policing Framework for Safer Cities and Communities Original Idea: 01 December 2019Author: Muhammad Asim – Global Progress VolunteerIndependent Researcher ID / ORCID: 0000-0002-8575-4447Relevant United Nations Sustainable Development Goal: SDG 11 – Sustainable Cities and Communities Abstract Street crime and urban insecurity remain important challenges for cities and communities worldwide. Although crime patterns differ considerably between countries and cities, effective public safety requires approaches that combine prevention, lawful enforcement, rehabilitation, community participation and evidence-based policing. This paper introduces the Camouflage Police Force (CPF), an original policing concept first developed by the author on 01 December 2019. CPF proposes an integrated three-stage framework: (1) root-cause prevention, (2) lawful surrender, rehabilitation and reintegration, and (3) adaptive camouflage policing. The distinctive element of CPF is its proposal to complement conventional visible policing with strategically less-obvious police deployment. As an initial research hypothesis, CPF proposes testing a 70% adaptive/non-obvious and 30% visible deployment model. This ratio is not presented as an established universal standard; rather, it should be experimentally evaluated and adjusted according to local crime patterns, legal requirements, operational capacity, community expectations and empirical results. The framework also proposes responsible use of artificial intelligence and data-driven tools for crime-pattern analysis and resource allocation, while emphasizing human oversight, privacy, due process, proportionality and independent accountability. CPF is aligned with the broader objective of United Nations Sustainable Development Goal 11, which calls for cities and human settlements to become inclusive, safe, resilient and sustainable [1]. It also reflects established evidence supporting multisectoral violence prevention, attention to risk factors, social reintegration and participatory urban safety strategies [2–6]. The paper concludes that CPF represents a researchable alternative to predominantly reactive policing by integrating prevention, rehabilitation and adaptive protection into a single framework. Keywords: Camouflage Police Force, CPF, street crime, urban safety, crime prevention, adaptive policing, hidden policing, rehabilitation, reintegration, artificial intelligence, SDG 11, sustainable cities, public safety.
Muhammad Asim - Global Progress Volunteer Muhammad Asim - Global Progress Volunteer· Zenodo (CERN European Organi...· 0 citations
