AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS最新文献

{"title":"AI-Driven Archaeometric Analysis of Golden Horde Copper Coins from Bolghar","authors":"R. Kh. Khramchenkova,&nbsp;D. G. Mukhametshin,&nbsp;A. A. Elizarov,&nbsp;A. G. Sitdikov,&nbsp;P. V. Fedan","doi":"10.3103/S0005105526700901","DOIUrl":"10.3103/S0005105526700901","url":null,"abstract":"<p>This pioneering study analyzes trace impurities in copper coins to identify medieval monetary circulation patterns. An interdisciplinary study of numismatic material from Bolghar, consisting of 98 copper coins, was conducted. This material includes 91 Golden Horde coins, 6 foreign coins of the same period, and 1 coin from the pre-Mongol period of the Zengid dynasty. The article includes attributions with images of the obverse and reverse of the coins, as well as archaeometric data from a chemical analysis of the monetary artifacts using two independent methods: X-ray fluorescence and emission spectral analysis. To date, this is the first domestic interdisciplinary study to cover a representative collection of copper coins from various rulers and mints, based on archaeological material from Bolghar. By studying the coins’ chemical composition, the impurity ratios in copper alloys were determined, making it possible to divide the collection into ten main groups, including coins from various mints and rulers of specific dates. Determining the impurity composition makes it possible not only to classify the coins into groups but also to formulate substantiated hypotheses regarding the geographic location of copper deposits used for minting. The resulting database, which combines numismatic attribution, coin images, and chemical composition data, has been published as an open dataset in the Zenodo repository and is designed to support the application of machine learning and artificial intelligence methods to the automatic identification and provenance analysis of medieval coins.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S548 - S557"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176549","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of Approaches to the Problem of Automatic Generation of Official Reply Letters Using LLMs","authors":"I. E. Nikolaev,&nbsp;A. V. Melnikov,&nbsp;K. E. Alekseev,&nbsp;A. S. Belonogov,&nbsp;M. A. Rusanov","doi":"10.3103/S0005105526700767","DOIUrl":"10.3103/S0005105526700767","url":null,"abstract":"<p>One of the key automation challenges for government agencies is the preparation of official response letters. This article presents an empirical comparison of two approaches to the automatic generation of official response letters: one based on templates defining the letter structure and one based on relevant letter examples selected using RAG. The original GovLetter dataset, which includes real-life business correspondence from the government agency Yugra in the Khanty-Mansi Autonomous Okrug was used as a basis. Generation was performed using a locally deployed open-source language model. The quality of the results was assessed according to 12 criteria using the Schema-Guided Reasoning (SGR) framework and the LLM-as-a-Judge methodology. The experimental results demonstrate that the example-based approach outperforms the template-based approach in most metrics, particularly in the accuracy of arguments, formal tone, and correct formatting. The obtained results confirm the potential of solutions based on document history for the effective automation of official response letter preparation.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S445 - S449"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Construction of a “Neuron Activity–Class” Mapping in a Convolutional Spiking Neural Network with STDP Training","authors":"A. S. Toshchev","doi":"10.3103/S0005105526700883","DOIUrl":"10.3103/S0005105526700883","url":null,"abstract":"<p>This paper investigates the problem of constructing a mapping between the spiking activity of a convolutional spiking neural network and the class of an input image. This problem arises after unsupervised training: the convolutional layer forms an event-based representation of the image, but the network itself does not contain a ready-made mechanism for assigning a class label. The aim of the study is to evaluate how informative the spike counters of the convolutional layer are after training with the STDP rule (Spike-Timing-Dependent Plasticity, a biologically plausible learning rule for impulse-based, or spiking, neural networks) and to assess the performance of a linear readout classifier as the number of training presentations increases. The experiments were conducted on the MNIST dataset, a standard dataset of handwritten digit images. A single-layer architecture was used: one convolutional layer with 32 feature maps, a 5 × 5 kernel, and an output dimensionality of 32 × 24 × 24, corresponding to 18432 LIF neurons, where LIF denotes the leaky integrate-and-fire neuron model. The input images were encoded using a deterministic Poisson encoder; the Poisson encoding multiplier was set to 0.006. The presentation duration, that is, the time during which an object was shown to the network, was 100 time steps. Training was performed from scratch using the STDP rule, without resuming from a previously saved state. For evaluation, a protocol was used in which spike counters were collected on 10000 training images and 10000 test images. In the main experiment, with 15 000 STDP training presentations, an accuracy of 0.8883 was achieved. The baseline label-map approach, close to the method of Diehl and Cook, produced an accuracy of about 0.48 under the same configuration. The coverage of the calibrated label map was 219 out of 18 432 neurons, while the average activity was 4146.9416 spikes per image. The neighboring scale point with 10 000 training presentations yielded an accuracy of 0.8876. The results show that the chosen method for reading out activity has a substantial effect on the final classification quality: a linear readout classifier based on the full vector of network spike counters extracts distributed information that is lost when labels are assigned directly to individual neurons.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S531 - S537"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Large Language Models for Word-In-Context","authors":"D. V. Kokosinskii","doi":"10.3103/S0005105526700743","DOIUrl":"10.3103/S0005105526700743","url":null,"abstract":"<p>Task-specific models have long dominated natural language processing tasks. However, general-purpose large language models (LLMs) have recently begun to successfully compete with highly specialized solutions across various NLP domains. This paper examines the applicability of LLMs to the problem of estimating the semantic proximity of word meanings in a pair of usages, known as Word-in-Context (WiC). Drawing on the multilingual CoMeDi benchmark, we propose novel approaches to building automated WiC systems based on LLMs. Five different configurations of LLM-based WiC systems were systematically compared in terms of their quality and computational cost. In particular, we propose a configuration where LLM predictions are adjusted using a training set, without the need to fine-tune the LLM itself. Results on test sets across seven languages show that our proposed approaches enable LLMs to outperform all existing specialized systems, establishing a new state-of-the-art (SOTA) on the CoMeDi benchmark. However, this higher quality comes at the cost of a significant rise in computing costs: LLM-based systems require several orders of magnitude more computations compared to compact, specialized models (such as XL-DURel). This work is one step toward understanding the trade-off between accuracy and resource efficiency when using modern LLMs in lexical semantics tasks.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S429 - S437"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176433","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Haruco: A New Russian-Language Corpus of Popular Science Texts with Coreference Annotation","authors":"R. D. Shuvalov,&nbsp;E. A. Sidorova","doi":"10.3103/S0005105526700822","DOIUrl":"10.3103/S0005105526700822","url":null,"abstract":"<p>This paper presents a new Russian-language corpus with coreference annotation, HaRuCo (Habr Russian Coreference Corpus). The corpus is based on popular science articles related to the subject area of Computational Linguistics. The paper proposes a method for coreference annotation in texts from narrow subject areas, which includes four main stages: syntactic analysis of the text, assembly of noun phrases and identification of pronouns for constructing mentions (spans), classification of mentions by subject area classes, and clustering of mentions according to chains of coreferentially linked spans. Coreferential links were annotated using a syntactic parser and a large language model, followed by manual verification and correction. The created corpus includes 3727 entities, 9905 mentions, and 2683 coreferential chains. The created corpus can be used for training and evaluating coreference resolution models for the Russian language.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S490 - S493"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic Speech Recognition Quality Prediction Based on Large Language Models","authors":"A. V. Polevoi","doi":"10.3103/S0005105526700779","DOIUrl":"10.3103/S0005105526700779","url":null,"abstract":"<p>Despite the rapid development in the field of automatic speech recognition (ASR) systems, the recognition quality remains poor for audio recordings with acoustic degradation (music, crowd shouts, sounds of machinery, etc.). This is particularly important when implementing models in critical areas that are characterized by increased attention to control and reliability of the results (aviation, medicine, autopilots, etc.). Error in such areas is very expensive and the use of models becomes impossible, even if the error occurs rarely. To reduce risks, it is advisable to evaluate the expected recognition quality in advance. This article proposes an approach to predicting the word error rate (WER) based on the acoustic characteristics of the signal and calculating the perplexity of language models. The proposed method involves the creation of diverse sets of audio data by applying various types of acoustic observations to pure speech samples at various levels of quality and intelligibility. Unlike previous studies, a complete set of speech features is extracted and analyzed: prediction of the value of the signal-to-noise ratio (SNR), neural network sound quality metrics (NISQA, etc.) as well as the perplexity of the text of the ASR hypothesis using the language model as an additional feature for training a unified model. Experiments are being conducted using modern speech recognition architectures to demonstrate the effectiveness of the proposed method in predicting WER in various acoustic conditions. It is shown that the inclusion of perplexity significantly improves the quality of the WER prediction, in particular for data where acoustic features are weakly correlated with recognition errors. The results are applicable to the automatic evaluation of the expected quality of speech recognition and filtering of audio inputs.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S450 - S458"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Graph-Based Semantic Analysis of a Scientific Article Corpus","authors":"V. A. Chunikhin,&nbsp;S. A. Zaitsev,&nbsp;O. M. Ataeva","doi":"10.3103/S0005105526700809","DOIUrl":"10.3103/S0005105526700809","url":null,"abstract":"<p>The problem of effective navigation and the search for relevant information in growing volumes of scientific publications requires a shift from classical full-text search methods to semantic models. This work proposes an approach to structuring a heterogeneous corpus of scientific texts by constructing a knowledge graph (KG). A data processing pipeline is developed that encompasses the extraction of metadata, keywords, and structural elements of articles, followed by their integration into a unified graph. Based on the constructed KG, methods for analyzing explicit and extracting implicit connections between publications are implemented. The research results demonstrate the effectiveness of the graph-based representation of scientific information for uncovering hidden patterns within subject domains and supporting intelligent navigation.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S474 - S479"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Experimental Validation of the Universality of a Modular Metadata Management Architecture for Heterogeneous Repositories","authors":"A. Bondyakov,&nbsp;A. Kondratyev","doi":"10.3103/S0005105526700913","DOIUrl":"10.3103/S0005105526700913","url":null,"abstract":"<p>The hypothesis regarding the independence of platform performance from its specific type when utilizing a developed adapter layer was experimentally validated. The study was conducted in an isolated Docker environment with fixed computational resource constraints (CPU, RAM, I/O), employing rigorous statistical analysis of the results. The obtained data revealed no statistically significant differences in processing times; the observed variability falls within the bounds of experimental error attributable to the characteristics of the Docker containerization environment. It is demonstrated that the overhead introduced by the proposed adapter layer constitutes less than 0.5% of the total processing time, thereby ensuring high efficiency and extensibility of the proposed architectural solution.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"230 - 239"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fuzzy-Logic Adaptation of Sliding Window Parameters in Data Preparation for Large Language Models","authors":"M. V. Bobyr,&nbsp;N. A. Milostnaya,&nbsp;S. Yu. Belskaia","doi":"10.3103/S0005105526700846","DOIUrl":"10.3103/S0005105526700846","url":null,"abstract":"<p>A fuzzy controller for calculating sliding window parameters is proposed to prepare training data for large language models. Traditional approaches set the “window step” and “context length” as fixed constants across the entire text without considering linguistic characteristics of individual fragments, such as rich scientific text and monotonous repeating text. The proposed method utilizes two automatically extracted features: lexical diversity and average BPE token length. Operating on the Mamdani inference algorithm with a nine-rule base and center-of-gravity defuzzification, the fuzzy controller adaptively calculates the “window step” and “context length” parameters for each fragment. The proposed approach offers a cognitive interpretation by mimicking human reading mechanics, where readers naturally apply adaptive attention; for example, complex fragments are processed more attentively using smaller incremental steps.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S499 - S506"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Framework for Safety Analysis of LLM-Generated Code in Multi-Agent Systems","authors":"D. A. Avagian,&nbsp;K. A. Ayrapetyants","doi":"10.3103/S000510552670072X","DOIUrl":"10.3103/S000510552670072X","url":null,"abstract":"<p>Large language models (LLMs) are increasingly being used for code generation. However, both the generated code and the underlying LLM-based systems demand rigorous security evaluation. A common approach to enhancing code generation quality involves multi-agent systems composed of multiple models. This paper evaluates the performance of GPT-OSS 20B, GPT-OSS 120B, and Qwen3-Coder 480B models in single-agent and multi-agent configurations, using two code security benchmarks, SecurityEval and CyberSecEval. The key contribution is SafeAICoder, an extensible and scalable framework for testing LLMs enabling distributed server-side generation of multi-module programs and tests, independent of client-side code.</p>","PeriodicalId":42995,"journal":{"name":"AUTOMATIC DOCUMENTATION AND MATHEMATICAL LINGUISTICS","volume":"60 4","pages":"S409 - S422"},"PeriodicalIF":0.4,"publicationDate":"2026-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149176382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}