Centre for Language and Semantic Technologies

Publications

• Book

  Edited by: A. Panchenko, D. Gubanov, M. Khachay et al.

  Analysis of Images, Social Networks and Texts. AIST 2024

  Iss. 2364. Springer Nature Switzerland, Cham, 2024.

• Article

  Чертоганов К. А.

  Немарковская семантическая диффузия в многомерных пространствах: подход на основе уравнений Маккина–Власова–Фоккера–Планка

  Information diffusion models traditionally focus on event propagation, user activity and network interactions, providing limited mathematical insight into the evolution of semantic content itself. In this paper, narrative evolution is formulated as the evolution of probability measures in a multidimensional semantic space. We construct a mathematical bridge from event cascades to empirical semantic measures and subsequently to nonlinear density dynamics. The resulting framework is described by a non-Markovian McKean--Vlasov--Fokker--Planck equation incorporating semantic drift, collective interaction and memory effects. Within this setting, stable narrative configurations are characterized as stationary density solutions, while structural semantic transitions are associated with changes in their stability properties. The proposed model provides a unified probabilistic description of semantic diffusion and establishes a theoretical foundation for the analysis of long-term narrative evolution, semantic stabilization and regime transitions in complex information environments.

  Вестник ЮУрГУ. Серия: Вычислительная математика и информатика. Челябинск.. 2026.

• Book chapter

  Chervyakov A., Isaeva U., Emelyanov A. et al.

  Multimodal Evaluation of Russian-language Architectures.

  Multimodal large language models (MLLMs) are currently at the center of research attention, showing rapid progress in scale and capabilities, yet their intelligence, limitations, and risks remain insufficiently understood. To address these issues, particularly in the context of the Russian language, where no multimodal benchmarks currently exist, we introduce MERA Multi, an open multimodal evaluation framework for Russian-spoken architectures. The benchmark is instruction-based and encompasses default text, image, audio, and video modalities, comprising 18 newly constructed evaluation tasks for both general-purpose models and modality-specific architectures (image-to-text, video-to-text, and audio-to-text). Our contributions include: (i) a universal taxonomy of multimodal abilities; (ii) 18 datasets created entirely from scratch with attention to Russian cultural and linguistic specificity, unified prompts, and metrics; (iii) baseline results for both closed-source and open-source models; (iv) a methodology for preventing benchmark leakage, including watermarking for private sets. While our current focus is on Russian, the proposed benchmark provides a replicable methodology for constructing multimodal benchmarks in typologically diverse languages, particularly within the Slavic language family.

  In bk.: Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). Vol. 1. Association for Computational Linguistics, 2026. P. 2114-2161.

• Working paper

  Меньшиков И. А., Бернадотт А. К., Elvimov N. S.

  Hessian-based lightweight neural network for brain vessel segmentation on a minimal training dataset

  Accurate segmentation of blood vessels in brain magnetic resonance angiography (MRA) is essential for successful surgical procedures, such as aneurysm repair or bypass surgery. Currently, annotation is primarily performed through manual segmentation or classical methods, such as the Frangi filter, which often lack sufficient accuracy. Neural networks have emerged as powerful tools for medical image segmentation, but their development depends on well-annotated training datasets. However, there is a notable lack of publicly available MRA datasets with detailed brain vessel annotations. To address this gap, we propose a novel semi-supervised learning lightweight neural network with Hessian matrices on board for 3D segmentation of complex structures such as tubular structures, which we named HessNet. The solution is a Hessian-based neural network with only 6000 parameters. HessNet can run on the CPU and significantly reduces the resource requirements for training neural networks. The accuracy of vessel segmentation on a minimal training dataset reaches state-of-the-art results. It helps us create a large, semi-manually annotated brain vessel dataset of brain MRA images based on the IXI dataset (annotated 200 images). Annotation was performed by three experts under the supervision of three neurovascular surgeons after applying HessNet. It provides high accuracy of vessel segmentation and allows experts to focus only on the most complex important cases. The dataset is available at https://git.scinalytics.com/terilat/VesselDatasetPartly.

  Statistical mechanics. arXie. arXive, 2025