arXiv subjects

Sergey I. Nikolenko

Publications and source records attributed to Sergey I. Nikolenko.

18 recordsLinked to original sources

arXiv2026

The Morphological Core of Dungan: A Two-Dialect Finite-State Model and a Multi-Genre Evaluation

Dungan, a Sinitic language of Central Asia written in a Cyrillic-based script, is described in detail in the grammatical literature, yet the quantitative properties of its morphology in actual usage have, to the best of our knowledge, never been measured systematically. This paper uses a finite-state morphological analyzer as a measuring instrument. Implemented with HFST and covering both dialect groups (the Gansu variety, which is the literary standard, and the Shaanxi variety), the model offers no new grammatical description; it formalises the knowledge accumulated in Dungan studies and makes it measurable on corpora of three genres. Three results follow. Overt inflection is rare and limited: only 9.3% of recognized tokens in the encyclopaedic register have an overt marker, the system has just ten categories, and degree marking is almost absent. Ambiguity is genuine but sharply localized: 78.1% of tokens receive a single analysis, and the residue sits almost entirely on two clitics, -di (genitive/progressive) and -ni (locative/prospective). And the grammatical core proves effectively closed, the claim the instrument is really needed for: between 78% and 95% of the tokens the analyzer fails on, depending on register, are simply absent from the lexicon, and the phenomena the model deliberately declines to implement account for at most 4.5% of those failures. Held-out coverage (80 to 85%) is no lower than development coverage (73%), while a stem list with no morphology already reaches 67.4%, so the morphology is worth 5.2 points. The open frontier of Dungan is lexical. The analyzer, its sources and every evaluation script are released openly.

cs.CL↗

arXiv2026

KyrgyzLLM-Bench: Benchmarking Kyrgyz Language Understanding

Evaluating large language models (LLMs) across languages remains challenging, as most multilingual benchmarks rely on translated English datasets, often obscuring linguistic and cultural specificity in the target language. This issue is particularly pronounced for less-resourced languages such as Kyrgyz, where reliable natively authored evaluation data are scarce. Building on previously introduced Kyrgyz-language evaluation datasets, this work reports the first systematic and large-scale evaluation of LLMs in Kyrgyz using the KyrgyzLLM-Bench benchmark suite. KyrgyzLLM-Bench comprises two natively authored datasets$-$KyrgyzMMLU and KyrgyzRC$-$together with carefully translated and manually post-edited versions of WinoGrande, HellaSwag, BoolQ, and TruthfulQA. We evaluate 26 open- and closed-source LLMs under zero-shot and few-shot settings, analyzing model performance, cross-lingual transfer, and the impact of translation artifacts on evaluation reliability. Across families and tasks, model rankings transfer broadly from English to Kyrgyz on WinoGrande and BoolQ, and to a lesser extent on MMLU, while HellaSwag exhibits a substantial English-Kyrgyz performance gap consistent with translation-induced plausibility shifts. Few-shot prompting improves several open-source models on reading comprehension but behaves inconsistently for proprietary models on translated tasks. We publicly release all datasets, evaluation code, and per-model results, and integrate the Kyrgyz tasks into a widely used multilingual evaluation framework to support future research on Kyrgyz NLP.

cs.CL↗

arXiv2026

Query2Diagram: Answering Developer Queries with UML Diagrams

Software documentation frequently becomes outdated or fails to exist entirely, yet developers need focused views of their codebase to understand complex systems. While automated reverse engineering tools can generate UML diagrams from code, they produce overwhelming detail without considering developer intent. We introduce query-driven UML diagram generation, where LLMs create diagrams that directly answer natural language questions about code. Unlike existing methods, our approach produces semantically focused diagrams containing only relevant elements with contextual descriptions. We fine-tune Qwen2.5-Coder-14B on a curated dataset of code files, developer queries, and corresponding diagram representations in a structured JSON format, evaluating with both automatic detection of structural defects and human assessment of semantic relevance. Results demonstrate that fine-tuning on a modest amount of manually corrected data yields dramatic improvements: our best model achieves the highest F1 scores while reducing defect rates below state-of-the-art LLMs, generating diagrams that are both structurally sound and semantically faithful to developer queries. Thus, we establish the feasibility of using LLMs for scalable contextual, on-demand documentation generation. We make our code and dataset publicly available at https://github.com/i-need-a-pencil/query2diagram.

cs.SE↗

arXiv2024

Syntactic Transfer to Kyrgyz Using the Treebank Translation Method

The Kyrgyz language, as a low-resource language, requires significant effort to create high-quality syntactic corpora. This study proposes an approach to simplify the development process of a syntactic corpus for Kyrgyz. We present a tool for transferring syntactic annotations from Turkish to Kyrgyz based on a treebank translation method. The effectiveness of the proposed tool was evaluated using the TueCL treebank. The results demonstrate that this approach achieves higher syntactic annotation accuracy compared to a monolingual model trained on the Kyrgyz KTMU treebank. Additionally, the study introduces a method for assessing the complexity of manual annotation for the resulting syntactic trees, contributing to further optimization of the annotation process.

cs.CL↗

arXiv2023

Benchmarking Multilabel Topic Classification in the Kyrgyz Language

Kyrgyz is a very underrepresented language in terms of modern natural language processing resources. In this work, we present a new public benchmark for topic classification in Kyrgyz, introducing a dataset based on collected and annotated data from the news site 24.KG and presenting several baseline models for news classification in the multilabel setting. We train and evaluate both classical statistical and neural models, reporting the scores, discussing the results, and proposing directions for future work.

cs.CL↗

arXiv2019

RecVAE: a New Variational Autoencoder for Top-N Recommendations with Implicit Feedback

Recent research has shown the advantages of using autoencoders based on deep neural networks for collaborative filtering. In particular, the recently proposed Mult-VAE model, which used the multinomial likelihood variational autoencoders, has shown excellent results for top-N recommendations. In this work, we propose the Recommender VAE (RecVAE) model that originates from our research on regularization techniques for variational autoencoders. RecVAE introduces several novel ideas to improve Mult-VAE, including a novel composite prior distribution for the latent codes, a new approach to setting the $β$ hyperparameter for the $β$-VAE framework, and a new approach to training based on alternating updates. In experimental evaluation, we show that RecVAE significantly outperforms previously proposed autoencoder-based models, including Mult-VAE and RaCT, across classical collaborative filtering datasets, and present a detailed ablation study to assess our new developments. Code and models are available at https://github.com/ilya-shenbin/RecVAE.

cs.IR↗

arXiv2019

Synthetic Data for Deep Learning

Synthetic data is an increasingly popular tool for training deep learning models, especially in computer vision but also in other areas. In this work, we attempt to provide a comprehensive survey of the various directions in the development and application of synthetic data. First, we discuss synthetic datasets for basic computer vision problems, both low-level (e.g., optical flow estimation) and high-level (e.g., semantic segmentation), synthetic environments and datasets for outdoor and urban scenes (autonomous driving), indoor scenes (indoor navigation), aerial navigation, simulation environments for robotics, applications of synthetic data outside computer vision (in neural programming, bioinformatics, NLP, and more); we also survey the work on improving synthetic data development and alternative ways to produce it such as GANs. Second, we discuss in detail the synthetic-to-real domain adaptation problem that inevitably arises in applications of synthetic data, including synthetic-to-real refinement with GAN-based models and domain adaptation at the feature/model level without explicit data transformations. Third, we turn to privacy-related applications of synthetic data and review the work on generating synthetic datasets with differential privacy guarantees. We conclude by highlighting the most promising directions for further work in synthetic data studies.

cs.LG↗

arXiv2019

New Competitiveness Bounds for the Shared Memory Switch

We consider one of the simplest and best known buffer management architectures: the shared memory switch with multiple output queues and uniform packets. It was one of the first models studied by competitive analysis, with the Longest Queue Drop (LQD) buffer management policy shown to be at least $\sqrt{2}$- and at most $2$-competitive; a general lower bound of $4/3$ has been proven for all deterministic online algorithms. Closing the gap between $\sqrt{2}$ and $2$ has remained an open problem in competitive analysis for more than a decade, with only marginal success in reducing the upper bound of $2$. In this work, we first present a simplified proof for the $\sqrt{2}$ lower bound for LQD and then, using a reduction to the continuous case, improve the general lower bound for all deterministic online algorithms from $\frac 43$ to $\sqrt{2}$. Then, we proceed to improve the lower bound of $\sqrt{2}$ specifically for LQD, showing that LQD is at least $1.44546086$-competitive. We are able to prove the bound by presenting an explicit construction of the optimal clairvoyant algorithm which then allows for two different ways to prove lower bounds: by direct computer simulations and by proving lower bounds via linear programming. The linear programming approach yields a lower bound for LQD of $1.4427902$ (still larger than $\sqrt{2}$).

cs.NI↗

arXiv2019

AspeRa: Aspect-based Rating Prediction Model

We propose a novel end-to-end Aspect-based Rating Prediction model (AspeRa) that estimates user rating based on review texts for the items and at the same time discovers coherent aspects of reviews that can be used to explain predictions or profile users. The AspeRa model uses max-margin losses for joint item and user embedding learning and a dual-headed architecture; it significantly outperforms recently proposed state-of-the-art models such as DeepCoNN, HFT, NARRE, and TransRev on two real world data sets of user reviews. With qualitative examination of the aspects and quantitative evaluation of rating prediction models based on these aspects, we show how aspect embeddings can be used in a recommender system.

cs.CL↗

arXiv2019

Adapting Convolutional Neural Networks for Geographical Domain Shift

We present the winning solution for the Inclusive Images Competition organized as part of the Conference on Neural Information Processing Systems (NeurIPS 2018) Competition Track. The competition was organized to study ways to cope with domain shift in image processing, specifically geographical shift: the training and two test sets in the competition had different geographical distributions. Our solution has proven to be relatively straightforward and simple: it is an ensemble of several CNNs where only the last layer is fine-tuned with the help of a small labeled set of tuning labels made available by the organizers. We believe that while domain shift remains a formidable problem, our approach opens up new possibilities for alleviating this problem in practice, where small labeled datasets from the target domain are usually either available or can be obtained and labeled cheaply.

cs.CV↗

arXiv2019

Label Denoising with Large Ensembles of Heterogeneous Neural Networks

Despite recent advances in computer vision based on various convolutional architectures, video understanding remains an important challenge. In this work, we present and discuss a top solution for the large-scale video classification (labeling) problem introduced as a Kaggle competition based on the YouTube-8M dataset. We show and compare different approaches to preprocessing, data augmentation, model architectures, and model combination. Our final model is based on a large ensemble of video- and frame-level models but fits into rather limiting hardware constraints. We apply an approach based on knowledge distillation to deal with noisy labels in the original dataset and the recently developed mixup technique to improve the basic models.

cs.CV↗

arXiv2019

SEIGAN: Towards Compositional Image Generation by Simultaneously Learning to Segment, Enhance, and Inpaint

We present a novel approach to image manipulation and understanding by simultaneously learning to segment object masks, paste objects to another background image, and remove them from original images. For this purpose, we develop a novel generative model for compositional image generation, SEIGAN (Segment-Enhance-Inpaint Generative Adversarial Network), which learns these three operations together in an adversarial architecture with additional cycle consistency losses. To train, SEIGAN needs only bounding box supervision and does not require pairing or ground truth masks. SEIGAN produces better generated images (evaluated by human assessors) than other approaches and produces high-quality segmentation masks, improving over other adversarially trained approaches and getting closer to the results of fully supervised training.

cs.CV↗

arXiv2019

Learning State Representations in Complex Systems with Multimodal Data

Representation learning becomes especially important for complex systems with multimodal data sources such as cameras or sensors. Recent advances in reinforcement learning and optimal control make it possible to design control algorithms on these latent representations, but the field still lacks a large-scale standard dataset for unified comparison. In this work, we present a large-scale dataset and evaluation framework for representation learning for the complex task of landing an airplane. We implement and compare several approaches to representation learning on this dataset in terms of the quality of simple supervised learning tasks and disentanglement scores. The resulting representations can be used for further tasks such as anomaly detection, optimal control, model-based reinforcement learning, and other applications.

cs.LG↗

arXiv2013

Balancing Work and Size with Bounded Buffers

We consider the fundamental problem of managing a bounded size queue buffer where traffic consists of packets of varying size, where each packet requires several rounds of processing before it can be transmitted from the queue buffer. The goal in such an environment is to maximize the overall size of packets that are successfully transmitted. This model is motivated by the ever-growing ubiquity of network processors architectures, which must deal with heterogeneously-sized traffic, with heterogeneous processing requirements. Our work addresses the tension between two conflicting algorithmic approaches in such settings: the tendency to favor packets with fewer processing requirements, thus leading to fast contributions to the accumulated throughput, as opposed to preferring packets of larger size, which imply a large increase in throughput at each step. We present a model for studying such systems, and present competitive algorithms whose performance depend on the maximum size a packet may have, and maximum amount of processing a packet may require. We further provide lower bounds on algorithms performance in such settings.

cs.NI↗

arXiv2012

BayesHammer: Bayesian clustering for error correction in single-cell sequencing

Error correction of sequenced reads remains a difficult task, especially in single-cell sequencing projects with extremely non-uniform coverage. While existing error correction tools designed for standard (multi-cell) sequencing data usually come up short in single-cell sequencing projects, algorithms actually used for single-cell error correction have been so far very simplistic. We introduce several novel algorithms based on Hamming graphs and Bayesian subclustering in our new error correction tool BayesHammer. While BayesHammer was designed for single-cell sequencing, we demonstrate that it also improves on existing error correction tools for multi-cell sequencing data while working much faster on real-life datasets. We benchmark BayesHammer on both $k$-mer counts and actual assembly results with the SPAdes genome assembler.

q-bio.QM↗

arXiv2012

FIFO Queueing Policies for Packets with Heterogeneous Processing

We consider the problem of managing a bounded size First-In-First-Out (FIFO) queue buffer, where each incoming unit-sized packet requires several rounds of processing before it can be transmitted out. Our objective is to maximize the total number of successfully transmitted packets. We consider both push-out (when the policy is permitted to drop already admitted packets) and non-push-out cases. In particular, we provide analytical guarantees for the throughput performance of our algorithms. We further conduct a comprehensive simulation study which experimentally validates the predicted theoretical behaviour.

cs.NI↗

arXiv2008

New Combinatorial Complete One-Way Functions

In 2003, Leonid A. Levin presented the idea of a combinatorial complete one-way function and a sketch of the proof that Tiling represents such a function. In this paper, we present two new one-way functions based on semi-Thue string rewriting systems and a version of the Post Correspondence Problem and prove their completeness. Besides, we present an alternative proof of Levin's result. We also discuss the properties a combinatorial problem should have in order to hold a complete one-way function.

cs.CC↗

arXiv2003

Hard satisfiable formulas for DPLL-type algorithms

We address lower bounds on the time complexity of algorithms solving the propositional satisfiability problem. Namely, we consider two DPLL-type algorithms, enhanced with the unit clause and pure literal heuristics. Exponential lower bounds for solving satisfiability on provably satisfiable formulas are proven.

cs.CC↗