Vascular tone plays a vital role in regulating blood pressure and coronary circulation, and it determines the peripheral vascular resistance. Vascular tone is dually regulated by the perivascular nerves and the cells in the inside lining of blood vessels (endothelial cells). Only a few methods for measuring vascular tone are available. Because of this, determining vascular tone in different arteries of the human body and monitoring tone changes is a vital challenge. This work presents an approach for determining vascular tone in human extremities based on multi-channel bioimpedance measurements. Detailed steps for processing the bioimpedance signals and extracting the main parameters from them have been presented. A graphical interface has been designed and implemented to display the vascular tone type in all channels with the phase of breathing during each cardiac cycle. This study is a key step towards understanding the way vascular tone changes in the extremities and how the nervous system regulates these changes. Future studies based on records of healthy and diseased people will contribute to increasing the possibility of early diagnosis of cardiovascular diseases.

We address the scarcity of datasets specifically tailored for legal NER in the Russian language and investigate the generalization capabilities of models towards unseen named entities. A rule-based program developed by legal experts at Tag-Consulting Company was employed to automatically annotate legal texts and create the RuLegalNER dataset. Part of the named entities only exists in the development and test splits, and they are unseen in the training set. RuBERT was utilized as the base architecture for experimental evaluation. Two different architectural extensions were explored: RuBERT with CRF and RuBERT with adapters. These architectures were used to train and evaluate NER models on the RuLegalNER dataset. Utilize RuLegalNER to train and evaluate legal NER models, enhancing performance in the legal domain and studying generalization on unseen entities. A published version of RuLegalNER is presented with detailed statistics and demonstration of the usefulness of RuLegalNER by evaluating modern architectures.

Large multi-label text classification is a challenging Natural Language Processing (NLP) problem that is concerned with text classification for datasets with thousands of labels. We tackle this problem in the legal domain, where datasets, such as JRC-Acquis and EURLEX57K labeled with the EuroVoc vocabulary were created within the legal information systems of the European Union. The EuroVoc taxonomy includes around 7000 concepts. In this work, we study the performance of various recent transformer-based models in combination with strategies such as generative pretraining, gradual unfreezing and discriminative learning rates in order to reach competitive classification performance, and present new state-of-the-art results of 0.661 (F1) for JRC-Acquis and 0.754 for EURLEX57K. Furthermore, we quantify the impact of individual steps, such as language model fine-tuning or gradual unfreezing in an ablation study, and provide reference dataset splits created with an iterative stratification algorithm.

Zero-shot cross-lingual transfer is an important feature in modern NLP models and architectures to support low-resource languages. In this work, We study zero-shot cross-lingual transfer from English to French and German under Multi-Label Text Classification, where we train a classifier using English training set, and we test using French and German test sets. We extend EURLEX57K dataset, the English dataset for topic classification of legal documents, with French and German official translation. We investigate the effect of using some training techniques, namely Gradual Unfreezing and Language Model finetuning, on the quality of zero-shot cross-lingual transfer. We find that Language model finetuning of multi-lingual pre-trained model (M-DistilBERT, M-BERT) leads to 32.0-34.94%, 76.15-87.54% relative improvement on French and German test sets correspondingly. Also, Gradual unfreezing of pre-trained model's layers during training results in relative improvement of 38-45% for French and 58-70% for German. Compared to training a model in Joint Training scheme using English, French and German training sets, zero-shot BERT-based classification model reaches 86% of the performance achieved by jointly-trained BERT-based classification model.

Generating coherent, grammatically correct, and meaningful text is very challenging, however, it is crucial to many modern NLP systems. So far, research has mostly focused on English language, for other languages both standardized datasets, as well as experiments with state-of-the-art models, are rare. In this work, we i) provide a novel reference dataset for Russian language modeling, ii) experiment with popular modern methods for text generation, namely variational autoencoders, and generative adversarial networks, which we trained on the new dataset. We evaluate the generated text regarding metrics such as perplexity, grammatical correctness and lexical diversity.