arXiv subjects

Ivan Smirnov

Publications and source records attributed to Ivan Smirnov.

At least 19 recordsLinked to original sources

arXiv2026

Modeling Stylistic Co-evolution in Symbolic Music Heritage Collections

Digitized musical heritage collections offer new opportunities to examine how stylistic traditions change over historical time, but computational analyses often reduce musical works to static classifications or similarity scores. This article proposes a representation-to-dynamics framework for studying cross-cultural harmonic change in Western art music. Starting from symbolic chord sequences, we derive contextual chord embeddings, project work-level representations into a shared harmonic space, and reconstruct country-level trajectories through temporally causal Kalman filtering. These trajectories are then modeled with DeGroot and Friedkin--Johnsen dynamics, yielding interpretable influence-like networks and estimates of stylistic anchoring. We apply the framework to a curated corpus of 480 dated works from 1875 to 1940 across Russia, France, Germany, Austria, and a heterogeneous ''Others'' group. Models fitted on 1875--1925 are evaluated through recursive forecasts for 1930--1940, testing whether the estimated dependency structure remains informative across a potentially changing historical and stylistic regime. The estimated trajectories and influence patterns are broadly consistent with established music-historical accounts of late Romantic and early modernist exchange, including the close relationship between German and Austrian traditions and historically plausible cross-currents between Russian and French traditions. PCA-variance-weighted estimation provides modest improvements while preserving a single interpretable influence network. Rather than treating the estimated matrices as direct causal evidence, the framework offers a reproducible quantitative layer for cultural heritage research, complementing archival and musicological interpretation.

cs.SD↗

arXiv2026

Structured Multi-Criteria Evaluation of Large Language Models with Fuzzy Analytic Hierarchy Process and DualJudge

Effective evaluation of large language models (LLMs) remains a critical bottleneck, as conventional direct scoring often yields inconsistent and opaque judgments. In this work, we adapt the Analytic Hierarchy Process (AHP) to LLM-based evaluation and, more importantly, propose a confidence-aware Fuzzy AHP (FAHP) extension that models epistemic uncertainty via triangular fuzzy numbers modulated by LLM-generated confidence scores. Systematically validated on JudgeBench, our structured approach decomposes assessments into explicit criteria and incorporates uncertainty-aware aggregation, producing more calibrated judgments. Extensive experiments demonstrate that both crisp and fuzzy AHP consistently outperform direct scoring across model scales and dataset splits, with FAHP showing superior stability in uncertain comparison scenarios. Building on these insights, we propose \textbf{DualJudge}, a hybrid framework inspired by Dual-Process Theory that adaptively fuses holistic direct scores with structured AHP outputs via consistency-aware weighting. DualJudge achieves state-of-the-art performance, underscoring the complementary strengths of intuitive and deliberative evaluation paradigms. These results establish uncertainty-aware structured reasoning as a principled pathway toward more reliable LLM assessment. Code is available at https://github.com/hreyulog/AHP_llm_judge.

cs.AI↗

arXiv2026

Automatic Classifiers Underdetect Emotions Expressed by Men

The widespread adoption of automatic sentiment and emotion classifiers makes it important to ensure that these tools perform reliably across different populations. Yet their reliability is typically assessed using benchmarks that rely on third-party annotators rather than the individuals experiencing the emotions themselves, potentially concealing systematic biases. In this paper, we use a unique, large-scale dataset of more than one million self-annotated posts and a pre-registered research design to investigate gender biases in emotion detection across 414 combinations of models and emotion-related classes. We find that across different types of automatic classifiers and various underlying emotions, error rates are consistently higher for texts authored by men compared to those authored by women. We quantify how this bias could affect results in downstream applications and show that current machine learning tools, including large language models, should be applied with caution when the gender composition of a sample is not known or variable. Our findings demonstrate that sentiment analysis is not yet a solved problem, especially in ensuring equitable model behaviour across demographic groups.

cs.CL↗

arXiv2025

RLBenchNet: The Right Network for the Right Reinforcement Learning Task

Reinforcement learning (RL) has seen significant advancements through the application of various neural network architectures. In this study, we systematically investigate the performance of several neural networks in RL tasks, including Long Short-Term Memory (LSTM), Multi-Layer Perceptron (MLP), Mamba/Mamba-2, Transformer-XL, Gated Transformer-XL, and Gated Recurrent Unit (GRU). Through comprehensive evaluation across continuous control, discrete decision-making, and memory-based environments, we identify architecture-specific strengths and limitations. Our results reveal that: (1) MLPs excel in fully observable continuous control tasks, providing an optimal balance of performance and efficiency; (2) recurrent architectures like LSTM and GRU offer robust performance in partially observable environments with moderate memory requirements; (3) Mamba models achieve a 4.5x higher throughput compared to LSTM and a 3.9x increase over GRU, all while maintaining comparable performance; and (4) only Transformer-XL, Gated Transformer-XL, and Mamba-2 successfully solve the most challenging memory-intensive tasks, with Mamba-2 requiring 8x less memory than Transformer-XL. These findings provide insights for researchers and practitioners, enabling more informed architecture selection based on specific task characteristics and computational constraints. Code is available at: https://github.com/SafeRL-Lab/RLBenchNet

cs.LG↗

arXiv2025

LAMeD: LLM-generated Annotations for Memory Leak Detection

Static analysis tools are widely used to detect software bugs and vulnerabilities but often struggle with scalability and efficiency in complex codebases. Traditional approaches rely on manually crafted annotations -- labeling functions as sources or sinks -- to track data flows, e.g., ensuring that allocated memory is eventually freed, and code analysis tools such as CodeQL, Infer, or Cooddy can use function specifications, but manual annotation is laborious and error-prone, especially for large or third-party libraries. We present LAMeD (LLM-generated Annotations for Memory leak Detection), a novel approach that leverages large language models (LLMs) to automatically generate function-specific annotations. When integrated with analyzers such as Cooddy, LAMeD significantly improves memory leak detection and reduces path explosion. We also suggest directions for extending LAMeD to broader code analysis.

cs.SE↗

arXiv2024

Navigating Ethical Challenges in Generative AI-Enhanced Research: The ETHICAL Framework for Responsible Generative AI Use

The rapid adoption of generative artificial intelligence (GenAI) in research presents both opportunities and ethical challenges that should be carefully navigated. Although GenAI tools can enhance research efficiency through automation of tasks such as literature review and data analysis, their use raises concerns about aspects such as data accuracy, privacy, bias, and research integrity. This paper develops the ETHICAL framework, which is a practical guide for responsible GenAI use in research. Employing a constructivist case study examining multiple GenAI tools in real research contexts, the framework consists of seven key principles: Examine policies and guidelines, Think about social impacts, Harness understanding of the technology, Indicate use, Critically engage with outputs, Access secure versions, and Look at user agreements. Applying these principles will enable researchers to uphold research integrity while leveraging GenAI benefits. The framework addresses a critical gap between awareness of ethical issues and practical action steps, providing researchers with concrete guidance for ethical GenAI integration. This work has implications for research practice, institutional policy development, and the broader academic community while adapting to an AI-enhanced research landscape. The ETHICAL framework can serve as a foundation for developing AI literacy in academic settings and promoting responsible innovation in research methodologies.

cs.CY↗

arXiv2024

Exploring Large Language Models for Detecting Mental Disorders

This paper compares the effectiveness of traditional machine learning methods, encoder-based models, and large language models (LLMs) on the task of detecting depression and anxiety. Five Russian-language datasets were considered, each differing in format and in the method used to define the target pathology class. We tested AutoML models based on linguistic features, several variations of encoder-based Transformers such as BERT, and state-of-the-art LLMs as pathology classification models. The results demonstrated that LLMs outperform traditional methods, particularly on noisy and small datasets where training examples vary significantly in text length and genre. However, psycholinguistic features and encoder-based models can achieve performance comparable to language models when trained on texts from individuals with clinically confirmed depression, highlighting their potential effectiveness in targeted clinical applications.

cs.CL↗

arXiv2024

Inference-Time Selective Debiasing to Enhance Fairness in Text Classification Models

We propose selective debiasing -- an inference-time safety mechanism designed to enhance the overall model quality in terms of prediction performance and fairness, especially in scenarios where retraining the model is impractical. The method draws inspiration from selective classification, where at inference time, predictions with low quality, as indicated by their uncertainty scores, are discarded. In our approach, we identify the potentially biased model predictions and, instead of discarding them, we remove bias from these predictions using LEACE -- a post-processing debiasing method. To select problematic predictions, we propose a bias quantification approach based on KL divergence, which achieves better results than standard uncertainty quantification methods. Experiments on text classification datasets with encoder-based classification models demonstrate that selective debiasing helps to reduce the performance gap between post-processing methods and debiasing techniques from the at-training and pre-processing categories.

cs.CL↗

arXiv2024

The Echoes of the 'I': Tracing Identity with Demographically Enhanced Word Embeddings

Identity is one of the most commonly studied constructs in social science. However, despite extensive theoretical work on identity, there remains a need for additional empirical data to validate and refine existing theories. This paper introduces a novel approach to studying identity by enhancing word embeddings with socio-demographic information. As a proof of concept, we demonstrate that our approach successfully reproduces and extends established findings regarding gendered self-views. Our methodology can be applied in a wide variety of settings, allowing researchers to tap into a vast pool of naturally occurring data, such as social media posts. Unlike similar methods already introduced in computer science, our approach allows for the study of differences between social groups. This could be particularly appealing to social scientists and may encourage the faster adoption of computational methods in the field.

cs.SI↗

arXiv2023

A Language Model for Grammatical Error Correction in L2 Russian

Grammatical error correction is one of the fundamental tasks in Natural Language Processing. For the Russian language, most of the spellcheckers available correct typos and other simple errors with high accuracy, but often fail when faced with non-native (L2) writing, since the latter contains errors that are not typical for native speakers. In this paper, we propose a pipeline involving a language model intended for correcting errors in L2 Russian writing. The language model proposed is trained on untagged texts of the Newspaper subcorpus of the Russian National Corpus, and the quality of the model is validated against the RULEC-GEC corpus.

cs.CL↗

arXiv2023

Light Coreference Resolution for Russian with Hierarchical Discourse Features

Coreference resolution is the task of identifying and grouping mentions referring to the same real-world entity. Previous neural models have mainly focused on learning span representations and pairwise scores for coreference decisions. However, current methods do not explicitly capture the referential choice in the hierarchical discourse, an important factor in coreference resolution. In this study, we propose a new approach that incorporates rhetorical information into neural coreference resolution models. We collect rhetorical features from automated discourse parses and examine their impact. As a base model, we implement an end-to-end span-based coreference resolver using a partially fine-tuned multilingual entity-aware language model LUKE. We evaluate our method on the RuCoCo-23 Shared Task for coreference resolution in Russian. Our best model employing rhetorical distance between mentions has ranked 1st on the development set (74.6% F1) and 2nd on the test set (73.3% F1) of the Shared Task. We hope that our work will inspire further research on incorporating discourse information in neural coreference resolution models.

cs.CL↗

arXiv2023

Toxic comments reduce the activity of volunteer editors on Wikipedia

Wikipedia is one of the most successful collaborative projects in history. It is the largest encyclopedia ever created, with millions of users worldwide relying on it as the first source of information as well as for fact-checking and in-depth research. As Wikipedia relies solely on the efforts of its volunteer-editors, its success might be particularly affected by toxic speech. In this paper, we analyze all 57 million comments made on user talk pages of 8.5 million editors across the six most active language editions of Wikipedia to study the potential impact of toxicity on editors' behaviour. We find that toxic comments consistently reduce the activity of editors, leading to an estimated loss of 0.5-2 active days per user in the short term. This amounts to multiple human-years of lost productivity when considering the number of active contributors to Wikipedia. The effects of toxic comments are even greater in the long term, as they significantly increase the risk of editors leaving the project altogether. Using an agent-based model, we demonstrate that toxicity attacks on Wikipedia have the potential to impede the progress of the entire project. Our results underscore the importance of mitigating toxic speech on collaborative platforms such as Wikipedia to ensure their continued success.

cs.CY↗

arXiv2023

The right to audit and power asymmetries in algorithm auditing

In this paper, we engage with and expand on the keynote talk about the Right to Audit given by Prof. Christian Sandvig at the IC2S2 2021 through a critical reflection on power asymmetries in the algorithm auditing field. We elaborate on the challenges and asymmetries mentioned by Sandvig - such as those related to legal issues and the disparity between early-career and senior researchers. We also contribute a discussion of the asymmetries that were not covered by Sandvig but that we find critically important: those related to other disparities between researchers, incentive structures related to the access to data from companies, targets of auditing and users and their rights. We also discuss the implications these asymmetries have for algorithm auditing research such as the Western-centrism and the lack of the diversity of perspectives. While we focus on the field of algorithm auditing specifically, we suggest some of the discussed asymmetries affect Computational Social Science more generally and need to be reflected on and addressed.

cs.CY↗

arXiv2023

A note on the constant characteristic time of failure incubation processes under various high-rate loads

The research reveals the existence of a constant characteristic time of preparatory micro-structural processes before the onset of macro-failure at various high loading rates of brittle and quasi-brittle materials. The presence of this characteristic is analysed based on available data in the literature from dynamic tests for uniaxial compression and splitting. It is shown that the characteristic time can be determined experimentally and used to calculate the strain rate dependencies of either critical failure stresses or time to failure, at least in the case of linearly growing loads. In addition, it is discussed that the presence of this constant parameter opens up a prospective opportunity for research and development of new methods for assessing the structural-temporal and scale characteristics of the strength and failure of materials under dynamic loads.

cond-mat.mtrl-sci↗

arXiv2021

Core but not peripheral online social ties is a protective factor against depression: evidence from a nationally representative sample of young adults

As social interactions are increasingly taking place in the digital environment, online friendship and its effects on various life outcomes from health to happiness attract growing research attention. In most studies, online ties are treated as representing a single type of relationship. However, our online friendship networks are not homogeneous and could include close connections, e.g. a partner, as well as people we have never met in person. In this paper, we investigate the potentially differential effects of online friendship ties on mental health. Using data from a Russian panel study (N = 4,400), we find that - consistently with previous research - the number of online friends correlates with depression symptoms. However, this is true only for networks that do not exceed Dunbar's number in size (N <= 150) and only for core but not peripheral nodes of a friendship network. The findings suggest that online friendship could encode different types of social relationships that should be treated separately while investigating the association between online social integration and life outcomes, in particular well-being or mental health.

cs.SI↗

arXiv2021

The FairCeptron: A Framework for Measuring Human Perceptions of Algorithmic Fairness

Measures of algorithmic fairness often do not account for human perceptions of fairness that can substantially vary between different sociodemographics and stakeholders. The FairCeptron framework is an approach for studying perceptions of fairness in algorithmic decision making such as in ranking or classification. It supports (i) studying human perceptions of fairness and (ii) comparing these human perceptions with measures of algorithmic fairness. The framework includes fairness scenario generation, fairness perception elicitation and fairness perception analysis. We demonstrate the FairCeptron framework by applying it to a hypothetical university admission context where we collect human perceptions of fairness in the presence of minorities. An implementation of the FairCeptron framework is openly available, and it can easily be adapted to study perceptions of algorithmic fairness in other application contexts. We hope our work paves the way towards elevating the role of studies of human fairness perceptions in the process of designing algorithmic decision making systems.

cs.CY↗

arXiv2020

A better method to enforce monotonic constraints in regression and classification trees

In this report we present two new ways of enforcing monotone constraints in regression and classification trees. One yields better results than the current LightGBM, and has a similar computation time. The other one yields even better results, but is much slower than the current LightGBM. We also propose a heuristic that takes into account that greedily splitting a tree by choosing a monotone split with respect to its immediate gain is far from optimal. Then, we compare the results with the current implementation of the constraints in the LightGBM library, using the well known Adult public dataset. Throughout the report, we mostly focus on the implementation of our methods that we made for the LightGBM library, even though they are general and could be implemented in any regression or classification tree. The best method we propose (a smarter way to split the tree coupled to a penalization of monotone splits) consistently beats the current implementation of LightGBM. With small or average trees, the loss reduction can be as high as 1% in the early stages of training and decreases to around 0.1% at the loss peak for the Adult dataset. The results would be even better with larger trees. In our experiments, we didn't do a lot of tuning of the regularization parameters, and we wouldn't be surprised to see that increasing the performance of our methods on test sets.

stat.ML↗

arXiv2020

Measuring Adolescents' Well-being: Correspondence of Naive Digital Traces to Survey Data

Digital traces are often used as a substitute for survey data. However, it is unclear whether and how digital traces actually correspond to the survey-based traits they purport to measure. This paper examines correlations between self-reports and digital trace proxies of depression, anxiety, mood, social integration and sleep among high school students. The study is based on a small but rich multilayer data set (N = 144). The data set contains mood and sleep measures, assessed daily over a 4-month period, along with survey measures at two points in time and information about online activity from VK, the most popular social networking site in Russia. Our analysis indicates that 1) the sentiments expressed in social media posts are correlated with depression; namely, adolescents with more severe symptoms of depression write more negative posts, 2) late-night posting indicates less sleep and poorer sleep quality, and 3) students who were nominated less often as somebody's friend in the survey have fewer friends on VK and their posts receive fewer "likes." However, these correlations are generally weak. These results demonstrate that digital traces can serve as useful supplements to, rather than substitutes for, survey data in studies on adolescents' well-being. These estimates of correlations between survey and digital trace data could provide useful guidelines for future research on the topic.

cs.SI↗