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. Cham: Springer, 2024.

• Article

  Seralan V., Srinivasan S., Anitha K. et al.

  Spatiotemporal dynamics in the bi-layer memristive Morris-Lecar neurons with different obstacles

  We report spiral-wave transitions in the memristive Morris-Lecar neuron model with differently structured obstacles
  in single and bilayer configurations. Wave propagation is initiated by applying an external periodic stimulus to the
  left-side nodes of the excitable media. During interactions with obstacles, rich spiral patterns emerge in both single-layer and bilayer configurations across various amplitudes and frequencies of the external stimulus, as well as for different diffusion coefficients. The spatio-energy is also calculated to demonstrate the existence of a spiral core within stable spiral waves. The statistical synchronization factor is computed over a range of diffusion coefficients to assess the presence of spiral waves and the absence of instabilities. Building on these single-layer dynamics, we then investigate multilayer effects by extending our study to a bilayer system, with one layer containing obstacles. We observe the transition of spiral waves between the layers across these various stimulus parameters, as well as for different intralayer coupling coefficients. Finally, the trend of synchronization between the two layers is investigated using the spatial interlayer synchronization error across a range of interlayer coupling strengths. The important findings of the present study are discussed in the conclusion.

  
  
  

  Chaos. 2026. Vol. 36. No. 10. P. 1-13.

• Book chapter

  Куделя А. В., Алшауи Р., Shirnin A.

  Lacuna Inc. at SemEval-2026 Task 4: Structurally Gated State-Space Models for Disentangling Narrative Similarity

  In this paper, we present the Invariant-Variant Disentangled State-Space Model (IVD-SSM),our submission to SemEval-2026 Task 4 on Narrative Story Similarity and Narrative Representation Learning. Evaluating narrative similarity is a profound computational challenge that requires models to look past concrete, superficial elements such as specific names, actors, objects, or settings to isolate and compareabstract patterns of causality and plot progression. To model these extended causal chainswithout the quadratic bottlenecks of standard Transformers, we leverage a hybrid State-SpaceModel (Jamba-1.5-Mini). Building upon this backbone, we introduce the Structurally Gated Alignment (SGA) head, a novel, differentiable algorithmic architecture. The SGA head operates on two scales: a heavily strided Macro-path maps the coarse structural skeleton of a story, which then acts as a gating mechanism to filter a full-resolution Micro-path, actively suppressing semantic noise and superficial keyword overlaps. Evaluated on both pairwisecomparative judgments (Track A) and dense representation learning (Track B), our approach demonstrates that explicitly disentangling structural invariants from lexical variants provides a robust, principled framework for deep narrative understanding.

  In bk.: Proceedings of the 20th International Workshop on Semantic Evaluation (2026). San Diego: Association for Computational Linguistics, 2026. P. 2347-2353.

• 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