National University of Kyiv-Mohyla Academy

To evaluate the incidence, timing, and risk factors for postoperative needling procedure and incisional bleb revision following XEN gel stent implantation in a large single-center cohort with a long-term follow-up. This retrospective observational cohort study included 773 eyes from 596 patients who underwent XEN gel stent implantation between December 2015 and July 2023. Patients had primary or secondary open-angle glaucoma and at least 12 months of follow-up. Primary outcomes were the need for surgical bleb revision (needling procedure or incisional bleb revision). Secondary outcomes included timing and frequency of interventions. Cox regression models were used to identify risk factors. A total of 417 eyes (53.9 What is new

Mixture-of-Experts (MoE) approaches have gained traction in robotics for their ability to dynamically allocate resources and specialize sub-networks. However, such systems typically rely on monolithic architectures with rigid, learned internal routing, which prevents selective expert customization and necessitates expensive joint training. We propose MoIRA, an architecture-agnostic modular framework that coordinates decoupled experts via an external, zero-shot text router. MoIRA employs two routing strategies: embedding-based similarity and prompt-driven language model inference. Leveraging Gr00t-N1 and pi 0 Vision-Language-Action models with low-rank adapters, we evaluate MoIRA on GR1 Humanoid tasks and LIBERO benchmarks. Our approach consistently outperforms generalist models and competes with fully trained MoE pipelines. Furthermore, we demonstrate system robustness against instruction perturbations. By relying on textual descriptions for zero-shot orchestration, MoIRA proves the viability of modular deployment and offers a scalable, flexible foundation for multi-expert robotic systems.

Artificial introgression hybridization within Triticeae is used to expand gene pools of cultivated species. Significant progress has been achieved, first of all, for traits of resistance to biotic stressors and product quality. The improvement of the methodical basis of molecular genetics now allows us to operate with the sequences of genome/transcriptome sequences and to compare hybrid genomes within the limits of their expression with the components of the original crosses. This changed the approach to evaluating the result of introgression. Traditionally, introgression was considered at the phenotypic level as a manifestation of the trait of interest and at the level of the nucleotide sequence of alien DNA present in the introgression line. In modern works, there is a gradual shift of attention from ascertaining the appearance of the target trait among the offspring to the study of the participation of epigenetic mechanisms in its implementation: DNA methylation and the participation of transposons and noncoding RNAs. These processes are activated by the very fact of combining two different genomes in one hybrid and originating an additional level of variability, which is reflected in the formation of the target phenotype. Currently, it is becoming obvious that changes at the level of nucleotide sequences and the organization of their expression can relate to both alien and genes of the recipient genome itself.

While biological neural networks develop from compact genomes using relatively simple rules, modern artificial neural architecture search methods mostly involve explicit and routine manual work. In this paper, we introduce MorphoNAS (Morphogenetic Neural Architecture Search), a system able to deterministically grow neural networks through morphogenetic self-organization inspired by the Free Energy Principle, reaction-diffusion systems, and gene regulatory networks. In MorphoNAS, simple genomes encode just morphogens dynamics and threshold-based rules of cellular development. Nevertheless, this leads to self-organization of a single progenitor cell into complex neural networks, while the entire process is built on local chemical interactions. Our evolutionary experiments focused on two different domains: structural targeting, in which MorphoNAS system was able to find fully successful genomes able to generate predefined random graph configurations (8-31 nodes); and functional performance on the CartPole control task achieving low complexity 6-7 neuron solutions when target network size minimization evolutionary pressure was applied. The evolutionary process successfully balanced between quality of of the final solutions and neural architecture search effectiveness. Overall, our findings suggest that the proposed MorphoNAS method is able to grow complex specific neural architectures, using simple developmental rules, which suggests a feasible biological route to adaptive and efficient neural architecture search.

У статті досліджено теоретико-методологічні засади синергетичної взаємодії цифрової економіки та людського капіталу в контексті сталого розвитку. Метою роботи є обґрунтування теоретико-прикладних засад та моделювання механізму синергії між цифровими технологіями та людським капіталом. Методологія базується на використанні комплексу наукових методів: системно-параметричного, абстрактно-логічного, економіко-статистичного. Авторами визначено поняття «цифровий людський капітал» як ключовий актив сучасної освітньої системи. Досліджено специфіку цифровізації освіти України, як чинника національної життєстійкості та інструмента повоєнного відновлення. Запропоновано методологічний підхід до оцінювання синергетичного ефекту на основі регресійного аналізу. Сформульовано практичні рекомендації щодо посилення взаємодії цифрових технологій та інтелектуального потенціалу для досягнення цілей сталого розвитку.