https://ai.elearning.sumdu.edu.ua Daily research digest on AI in education and technical countermeasures https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2609.15864 Durable skills, such as collaboration, creativity and critical thinking, are instrumental to success in the modern workforce. Yet, measuring these skills remains a persistent challenge. Moreover, because what is not measured is often not taught, these skills are often overlooked in mainstream educat https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2609.15864 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2609.17883 The rapid adoption of generative AI has made final artifacts unreliable evidence of student learning, and AI detectors that examine only the finished product are inaccurate and ethically contentious. Process data offers an alternative, but prior work covers only English essay writing. We ask whether https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2609.17883 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2605.16041 Machine learning systems increasingly make life-changing decisions about individuals, such as loan approvals, hiring, and cheating detection, raising a pressing question: how can individuals respond to negative decisions made by these opaque systems? While explainable artificial intelligence (XAI) h https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2605.16041 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2608.11256 Institutions use commercial AI detectors for academic integrity, yet detectors cannot distinguish AI editing from full LLM drafts and may treat both as misconduct. In a controlled study of published English abstracts (four domains; 2013 to 2015 vs. 2023 to 2025), we quantify this policy failure unde https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2608.11256 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2606.15643 Multilingual benchmarks are central to evaluating large language models (LLMs) across languages, but they suffer from three issues: exhaustive evaluation scales linearly with the number of languages, automatic translation introduces errors that are easily missed at scale, and some items conflate gen https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2606.15643 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2607.29252 Reliable evaluation of open-ended LLM outputs requires fine-grained rubrics, yet expert curation is costly and difficult to scale. Existing automated pipelines rely on strict judge unanimity and binary variance filters, which cannot distinguish measurable rubrics from informative ones. We introduce https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2607.29252 https://ai.elearning.sumdu.edu.ua/uk/doi/10.5281/zenodo.22998204 Public leaderboards order AI models by a point score and the public debate about them turns on leads of one or two points. We ask how many of those leads the benchmarks can detect. Using only the item-level results that the benchmarks themselves publish, we computed a psychometric report card for 61 https://ai.elearning.sumdu.edu.ua/uk/doi/10.5281/zenodo.22998204 https://ai.elearning.sumdu.edu.ua/uk/doi/10.1111/bmsp.70076 Abstract Measurement non‐invariance arises when the psychometric properties of a scale differ across subgroups, undermining the validity of group comparisons. At the item level, this manifests as differential item functioning (DIF), where item responses differ across groups after controlling for the https://ai.elearning.sumdu.edu.ua/uk/doi/10.1111/bmsp.70076 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2609.03215 Writing proficiency manifests in how students develop content, organize ideas, choose words, and use language. Despite growing interest in LLM-based student simulation, whether LLMs can reproduce such multidimensional variation in extended writing remains largely unexplored. In this work, we explore https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2609.03215 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2606.05180 Automated scoring models are increasingly used to assign rubric-based quality ratings to complex language performances, including classroom transcripts, yet they typically provide little insight into why a particular score is produced. We propose a general framework for sentence-level interpretabili https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2606.05180 https://ai.elearning.sumdu.edu.ua/uk/doi/10.5281/zenodo.21039771 The emergence of Large Language Models (LLMs) has renewed interest in automated essay scoring, yet a barrier to adoption persists: LLM-generated scores are sensitive to prompt formulation, and it is unclear which aspects of a prompt drive this effect or whether findings from one model or institution https://ai.elearning.sumdu.edu.ua/uk/doi/10.5281/zenodo.21039771 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.31041 Generative artificial intelligence promises to democratize personalized tutoring, yet the longitudinal stability of adaptive systems remains under-theorized. Current ethical discourse focuses on static bias. This paper argues that a distinct threat lies in algorithmic drift: a feedback loop in which https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.31041 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2606.00093 Whether a rubric-based LLM judge can replace human annotation is decided by its measured agreement with human labels. Yet the same verdicts can support wildly varying agreement numbers, depending on seemingly minor choices: the judgment scale, the retained cases, the handling of abstentions and inva https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2606.00093 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.25724 Automated essay scoring (AES) using large language models (LLMs) offers promising use cases for language assessment, yet high-stakes testing requires scores that are both controllable and verifiable, which current LLM-based scorers lack. Given that, we investigate whether LLM-based scorers form dist https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.25724 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2605.30504 LLM benchmark labels are frozen at release and silently propagated into downstream benchmarks, errors and all. We introduce an Item Response Theory-based indicator that surfaces likely mislabels at 95% precision in the top 200 examples across seven preference and multiple-choice benchmarks using res https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2605.30504 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2504.19306 High-dimensional changepoint inference, adaptable to diverse alternative scenarios, has attracted significant attention in recent years. In this paper, we propose an adaptive and robust approach to changepoint testing. Specifically, by generalizing the classical mean-based cumulative sum (CUSUM) sta https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2504.19306 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.06170 Biomedical large language model (LLM) evaluation requires auditable assessment of narrow, evolving, source-grounded subspecialty knowledge. Multiple sclerosis MRI (MS-MRI) provides a high-stakes textual-knowledge test case because correct reasoning requires current diagnostic criteria, standardized https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.06170 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.03273 Reinforcement learning (RL) for learning path recommendation (LPR) faces two coupled obstacles. First, the policy must commit to a sequence of L concepts without intermediate feedback, producing a combinatorial search space that grows super-exponentially with L and provides reward only at the final https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.03273 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.04125 Self-report is an appealing low-cost probe of an LLM's dispositions, but recent work finds only selective agreement between what models report and how they behave. Prior accounts establish these patterns by prompting black-box LLMs, leaving open whether the gap is a prompting artefact or a fact abou https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.04125 https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.01627 Difficulty is one of the most fundamental properties of a question: it determines whether the question can meaningfully discriminate between models of differing ability. Although a variety of methods can now estimate or predict difficulty automatically, they yield only a single descriptive number, w https://ai.elearning.sumdu.edu.ua/uk/doi/10.48550/arxiv.2610.01627