Association for Language Learning

Nicotinamide adenine dinucleotide (NAD⁺) declines with age, motivating "NAD⁺-boosting" strategies ranging from lifestyle interventions to supplementation with NAD⁺ precursors (e.g., nicotinamide riboside [NR], nicotinamide mononucleotide [NMN]) and, in some wellness settings, parenteral NAD⁺ administration. We conducted a PRISMA-guided systematic review of peer-reviewed human and rodent intervention studies (January 2010-October 2025) evaluating NAD-related compounds administered orally or parenterally. We identified 113 eligible studies: 33 human intervention studies (28 randomized; 5 nonrandomized) and 80 rodent studies. In rodent models, NAD⁺ augmentation was frequently associated with improvements in metabolic, mitochondrial, inflammatory, and functional outcomes, although effects varied across models and endpoints. In humans, oral NR and NMN consistently demonstrated biochemical target engagement (circulating (plasma/whole blood) or cellular (e.g., PBMC) NAD-related metabolites) and were generally well tolerated over weeks to months; however, effects on functional, metabolic, vascular, and other healthspan-relevant outcomes were heterogeneous and often null or endpoint-specific. No eligible outcomes trials evaluated intravenous or intramuscular NAD⁺ itself for anti-aging or wellness indications. One nonrandomized intravenous NMN study met inclusion criteria and primarily contributed short-term safety and biomarker information. An intravenous NAD⁺ pharmacokinetic pilot lacking eligible clinical outcomes was identified as contextual evidence only. Overall, NAD⁺ augmentation shows clear biological activity, but clinical effectiveness for anti-aging or wellness outcomes remains inconclusive. Larger, well-designed randomized trials with longer follow-up and prespecified clinically meaningful endpoints are needed, particularly for parenteral approaches.

Many organizations are increasingly relying on unstructured documents such as PDFs and scanned forms to support downstream large language model (LLM) services, including search, summarization, and recommendation. However, traditional OCR systems struggle with diverse layouts of documents, leading to frequent errors and high costs of labor. So, this study developed DATALUX - a robust document layout system that trans-forms unstructured documents into structured, machine-readable data suitable for automation. Built on a trans-former-based detector, DATALUX incorporates several modules for layout refinement, text-visual fusion, and layer-wise optimization to improve coherence and generalization across diverse layouts. Around January 2025, we successfully deployed DATALUX into one of the largest academic content service firms (Nurimedia) in South Korea. This firm faced the challenge of extracting metadata and references from thousands of academic pa-pers submitted in various formats. Also, the existing LLM-based tools provided unreliable results. So, they needed to process them manually, creating bottlenecks in both labor and time. However, DATALUX enabled the automatic structuring of over 100,000 research papers a year, improving extraction accuracy to over 97%, reducing costs by more than USD 185K annually, and accelerating processing speed by 8.7 times. These deployment results suggest that DATALUX enables scalable and efficient document automation in complex and high-volume environments successfully. We thus believe that our DATALUX has a significant impact on both academia and industry practices.

Science today operates in a rapidly changing environment, where more traditional notions of academic excellence are being challenged by new expectations for openness, impact, and accountability. This panel will explore several dimensions of this shift shaping the future of research and scholarly communication. First, we ask what “excellence” should mean in the 21st century and how that concept is evolving across disciplines and contexts. Second, we examine research culture and the design of reward systems: how incentives can foster integrity, collaboration, and societal relevance, and what accountability looks like for stakeholders in the system. Finally, we turn to the dissemination of research, considering what role peer review and a consideration of audiences can and should play in the evolution of academic publishing.

IntroductionThe integration of advanced digital applications within the educational sector remains a significant challenge. It has been determined that extracurricular educational settings possess a considerable capacity to enhance and differentiate digital instruction. However, educators frequently require assistance to transition from the use of existing digital resources to the creation of their own, personalized, customized learning materials. Professional learning communities (PLCs) have been identified as a promising framework in which teachers can collaborate to develop innovative and practical concepts. While the potential of PLCs is widely recognized, research on this topic is fragmented. A multitude of definitions and operationalizations have been proposed regarding the structure and functioning of PLCs. This complicates a systematic investigation of their effectiveness and makes it difficult to distinguish empirically based findings from purely anecdotal reports. Consequently, a definitive evidence base for the design of effective PLCs remains elusive.MethodsThe present study posits that the afore mentioned gap is addressed by means of a scoping review. Utilizing the PRISMA-ScR framework, a systematic analysis of German and English-language literature is conducted with the objective of mapping and synthesizing research on the composition, collaboration, and effectiveness of PLCs in the STEM field or in relation to digitalization.ResultsThe results of the study provide a structured overview of the key characteristics, structure, and success factors of PLCs. The present article elucidates the findings that emerged from a comparative and harmonising analysis of diverging PLCs.DiscussionThe ensuing discussion pertains to the configuration, efficacious design, and execution of professional learning communities, in conjunction with their operational methodologies and investigative endeavours. This synthesis provides a foundation for the conceptualisation of future cooperative initiatives aimed at the development of STEM- or digital educational resources.

Jinty Nelson left a strong mark of her presence in every institution that she was ever a part of, and there is none which does not regard its association with her as a source of deep pride.Past and Present is no exception.She was an immensely valued member of the Editorial Board for nearly twenty years, starting in 1995 and leaving in 2013, when she reached the mandatory retirement age of seventy.Jinty studied at Newnham College, Cambridge, where she was a star student, made the University Challenge team, and later wrote her Ph.D. dissertation under the direction of Walter Ullmann. 1 Ullmann did not approve of the way the thesis turned out, and their relationship was a fraught one, though it was eventually repaired.When discussing him later in life, she would sometimes quote what he had reportedly fumed to someone praising her work, after

The majority of mutational signatures have been characterized in tumors from American and European countries, and the degree to which mutational signatures are similar to or different from those in Chinese populations has not been fully explored. We leveraged a large-scale clinical sequencing cohort of tumors from the Chinese population developed by OrigiMed (OM). A total of 2,115 tumors from 25 major tumor types that had at least 10 single-base substitutions (SBS) were used for mutational signature analysis. Sixteen mutational signatures from the Catalogue of Somatic Mutations in Cancer (COSMIC) database were identified in the OM cohort using the musicatk package. The activity levels for most of the signatures were similar to those in a cohort from Memorial Sloan Kettering in the United States. The activity level for the apolipoprotein B mRNA editing catalytic polypeptide-like–related signature SBS2 was significantly lower in breast tumors in the OM cohort. The aristolochic acid signature SBS22 was observed in four soft-tissue sarcomas, and the POLE-associated signature SBS10 was observed in a gallbladder carcinoma. In lung adenocarcinoma, the polycyclic aromatic hydrocarbon signature SBS4 was significantly higher in males compared with females but not associated with smoking status. UV-related mutations were significantly lower in cutaneous melanomas from the Chinese cohort compared with the American cohort and may contribute to lower response rates to immunotherapy observed in this population. Overall, these results add to our understanding of the mutational processes that contribute to tumors from the Chinese population.Significance:Analysis of a large Chinese cohort from 25 tumor types reveals similarities and differences in the activity of mutational signatures compared with other populations.

The bridge revolution: Mind sports reshape Scotland’s educational landscape Mind sports education is gaining momentum in schools. Researchers at the University of Stirling are investigating the ways that strategic games like bridge can enhance skills development and academic engagement. As Scotland overhauls its Curriculum for Excellence and reimagines its assessment practices following the 2024 Independent Review of Qualifications and Assessment, could a deck of cards prove to be an unexpected yet valuable tool for educational innovation? Armed with only a deck of cards, educators show that classic card games like bridge could lead the way to curriculum enhancement.

x

Best-of-<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$n$</tex> (BoN) sampling is a practical approach for aligning language model outputs with human preferences without expensive fine-tuning. BoN sampling is performed by generating <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$n$</tex> responses to a prompt and then selecting the sample that maximizes a reward function. BoN yields high reward values in practice at a distortion cost, as measured by the KL-divergence between the sampled and original distribution. This distortion is coarsely controlled by varying the number of samples: larger <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$n$</tex> yields a higher reward at a higher distortion cost. We introduce Soft Best-of-<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$n$</tex> sampling, a generalization of BoN that allows for smooth interpolation between the original distribution and reward-maximizing distribution through a temperature parameter A. We establish theoretical guarantees showing that Soft Best-of-n sampling converges sharply to the optimal tilted distribution at a rate of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$O(1/n)$</tex> in KL and the expected (relative) reward. For sequences of discrete outputs, we analyze an additive reward model that reveals the fundamental limitations of blockwise sampling.