arXiv subjects

Vladimir Ivanov

Publications and source records attributed to Vladimir Ivanov.

At least 19 recordsLinked to original sources

arXiv2026

Recontextualization Mitigates Specification Gaming without Modifying the Specification

Developers often struggle to specify correct training labels and rewards. Perhaps they don't need to. We propose recontextualization, which reduces how often language models "game" training signals, performing misbehaviors those signals mistakenly reinforce. We show recontextualization prevents models from learning to 1) prioritize evaluation metrics over chat response quality; 2) special-case code to pass incorrect tests; 3) overwrite evaluation functions rather than write correct code; and 4) become sycophantic. Our method works by generating completions from prompts discouraging misbehavior and then recontextualizing them as though they were in response to prompts permitting misbehavior. Recontextualization trains language models to resist misbehavior even when instructions permit it. This mitigates the reinforcement of misbehavior from misspecified training signals, reducing specification gaming without improving the supervision signal.

cs.AI↗

arXiv2026

Training LLMs with Fault Tolerant HSDP on 100,000 GPUs

Large-scale training systems typically use synchronous training, requiring all GPUs to be healthy simultaneously. In our experience training on O(100K) GPUs, synchronous training results in a low efficiency due to frequent failures and long recovery time. To address this problem, we propose a novel training paradigm, Fault Tolerant Hybrid-Shared Data Parallelism (FT-HSDP). FT-HSDP uses data parallel replicas as units of fault tolerance. When failures occur, only a single data-parallel replica containing the failed GPU or server is taken offline and restarted, while the other replicas continue training. To realize this idea at scale, FT-HSDP incorporates several techniques: 1) We introduce a Fault Tolerant All Reduce (FTAR) protocol for gradient exchange across data parallel replicas. FTAR relies on the CPU to drive the complex control logic for tasks like adding or removing participants dynamically, and relies on GPU to perform data transfer for best performance. 2) We introduce a non-blocking catch-up protocol, allowing a recovering replica to join training with minimal stall. Compared with fully synchronous training at O(100K) GPUs, FT-HSDP can reduce the stall time due to failure recovery from 10 minutes to 3 minutes, increasing effective training time from 44\% to 80\%. We further demonstrate that FT-HSDP's asynchronous recovery does not bring any meaning degradation to the accuracy of the result model.

cs.DC↗

arXiv2025

Separation of Hoke and Schottky effects for improvement of mixed halide perovskite solar cell stability

Ion migration in halide perovskites is a key factor limiting the operational stability of solar cells due to formation of halogen ion enriched domains and accumulation layers. The present work demonstrates the manifestation of ion migration in two ways via Hoke and Schottky effects. Both effects are induced by external exposure but have its peculiar way of solar cell performance degradation. We demonstrate the effects of ion migration on the device performance by measuring time dependent short-circuit current and different impedance characteristics that allow to see how charge-carrier separation property degrades. The Schottky effect leads to the rapid decrease of charge-carrier separation characteristic of solar cell while Hoke effect leads to the slow defect accumulation in the perovskite layer leading to the enhanced Shockley-Read-Hall recombination. Separation of these two effects can be realized by simple increase of transport layer thickness. A thick transport layer blocks losses of charge-carrier selectivity in a solar cell and leads to an enormous increase of T80 time, from 15 seconds up to 60 minutes.

cond-mat.mtrl-sci↗

arXiv2025

Specifics of ITO properties deposited on cerium-doped glass for space-grade solar cells

Ce-doped glass is a well-established solution for ultraviolet and ionizing radiation shielding of solar cells in space. Traditionally, Ce-glass protected Si or III-V based devices as an overlaying cap. However, for emerging photovoltaics such as halide perovskites, thin Ce-glass coated with transparent conductive layers could serve as a lightweight carrier with an electrode. While indium-tin oxide (ITO) is widely used in solar cells for charge collection, its optical, structural, and electrical properties depend on the substrate quality. In this work, we demonstrated significant differences in properties of ITO deposited on Ce-glass (100 micron thick) compared to standard soda lime glass. ITO on Ce-glass exhibited pronounced compressive strain because of higher oxygen vacancy concentrations, reduced transparency and charge carrier concentration (10^19 cm-3) resulting from altered stoichiometry. Electrical analysis showed increased Hall mobility (66 cm2V-1s-1) but decreased conductivity due to excess tin incorporation. These specific ITO features originated from inelastic collisions with the substrate during deposition. Variations in wettability and surface potential underscore substrate-induced differences critical for developing optimized ITO coatings for space-grade photovoltaics.

cond-mat.mtrl-sci↗

arXiv2025

YABLoCo: Yet Another Benchmark for Long Context Code Generation

Large Language Models demonstrate the ability to solve various programming tasks, including code generation. Typically, the performance of LLMs is measured on benchmarks with small or medium-sized context windows of thousands of lines of code. At the same time, in real-world software projects, repositories can span up to millions of LoC. This paper closes this gap by contributing to the long context code generation benchmark (YABLoCo). The benchmark featured a test set of 215 functions selected from four large repositories with thousands of functions. The dataset contained metadata of functions, contexts of the functions with different levels of dependencies, docstrings, functions bodies, and call graphs for each repository. This paper presents three key aspects of the contribution. First, the benchmark aims at function body generation in large repositories in C and C++, two languages not covered by previous benchmarks. Second, the benchmark contains large repositories from 200K to 2,000K LoC. Third, we contribute a scalable evaluation pipeline for efficient computing of the target metrics and a tool for visual analysis of generated code. Overall, these three aspects allow for evaluating code generation in large repositories in C and C++.

cs.CL↗

arXiv2025

Stabilization of interfaces for double-cation halide perovskites with AVA2FAPb2I7 additives

The use of mixed cation absorber composition was considered as an efficient strategy to mitigate the degradation effects in halide perovskite solar cells. Despite the reports about partial stabilization at elevated temperatures, unfavorable phase transition after thermocycling and electric field-driven corrosion remains critical bottlenecks of perovskite thin-films semiconductors. In this work, we developed stabilized heterostructures based on CsFAPbI3, modified with mechanically synthesized quasi-2D perovskite incorporating the 5-ammonium valeric acid cation (AVA2FAPb2I7). We found that integration of AVA2FAPb2I7 into grain boundaries boosts phase resilience under harsh thermocycling from -10 up to 100 C and suppresses transitions, as well as decomposition to PbI2. The rapid oxidation of metal contacts in the multi-layer stacks with non-passivated CsFAPbI3 was effectively suppressed in the fabricated heterostructure. A comprehensive interface study of the copper electrode contact revealed that the incorporation of AVA2FAPb2I7 stabilized the lead and iodine states and suppressed contamination of FA cation in ambient conditions. Meanwhile, the metal-perovskite interface remained predominantly in the Cu(0)-Cu(I) state. The observed stabilization in perovskite heterostructure was attributed to an increased activation energy for delta-phase accumulation at the grain boundaries combined with reduced ionic diffusion. The obtained results opened important highlights for the mechanisms of the improved phase stability after thermal cycling and mitigation of the interface corrosion and under an applied electric field.

cond-mat.mtrl-sci↗

arXiv2025

Code Summarization Beyond Function Level

Code summarization is a critical task in natural language processing and software engineering, which aims to generate concise descriptions of source code. Recent advancements have improved the quality of these summaries, enhancing code readability and maintainability. However, the content of a repository or a class has not been considered in function code summarization. This study investigated the effectiveness of code summarization models beyond the function level, exploring the impact of class and repository contexts on the summary quality. The study involved revising benchmarks for evaluating models at class and repository levels, assessing baseline models, and evaluating LLMs with in-context learning to determine the enhancement of summary quality with additional context. The findings revealed that the fine-tuned state-of-the-art CodeT5+ base model excelled in code summarization, while incorporating few-shot learning and retrieved code chunks from RAG significantly enhanced the performance of LLMs in this task. Notably, the Deepseek Coder 1.3B and Starcoder2 15B models demonstrated substantial improvements in metrics such as BLEURT, METEOR, and BLEU-4 at both class and repository levels. Repository-level summarization exhibited promising potential but necessitates significant computational resources and gains from the inclusion of structured context. Lastly, we employed the recent SIDE code summarization metric in our evaluation. This study contributes to refining strategies for prompt engineering, few-shot learning, and RAG, addressing gaps in benchmarks for code summarization at various levels. Finally, we publish all study details, code, datasets, and results of evaluation in the GitHub repository available at https://github.com/kilimanj4r0/code-summarization-beyond-function-level.

cs.CL↗

arXiv2024

Leveraging Large Language Models in Code Question Answering: Baselines and Issues

Question answering over source code provides software engineers and project managers with helpful information about the implemented features of a software product. This paper presents a work devoted to using large language models for question answering over source code in Python. The proposed method for implementing a source code question answering system involves fine-tuning a large language model on a unified dataset of questions and answers for Python code. To achieve the highest quality answers, we tested various models trained on datasets preprocessed in different ways: a dataset without grammar correction, a dataset with grammar correction, and a dataset augmented with the generated summaries. The model answers were also analyzed for errors manually. We report BLEU-4, BERTScore F1, BLEURT, and Exact Match metric values, along with the conclusions from the manual error analysis. The obtained experimental results highlight the current problems of the research area, such as poor quality of the public genuine question-answering datasets. In addition, the findings include the positive effect of the grammar correction of the training data on the testing metric values. The addressed findings and issues could be important for other researchers who attempt to improve the quality of source code question answering solutions. The training and evaluation code is publicly available at https://github.com/IU-AES-AI4Code/CodeQuestionAnswering.

cs.CL↗

arXiv2024

Towards Safe Multilingual Frontier AI

Linguistically inclusive LLMs -- which maintain good performance regardless of the language with which they are prompted -- are necessary for the diffusion of AI benefits around the world. Multilingual jailbreaks that rely on language translation to evade safety measures undermine the safe and inclusive deployment of AI systems. We provide policy recommendations to enhance the multilingual capabilities of AI while mitigating the risks of multilingual jailbreaks. We examine how a language's level of resourcing relates to how vulnerable LLMs are to multilingual jailbreaks in that language. We do this by testing five advanced AI models across 24 official languages of the EU. Building on prior research, we propose policy actions that align with the EU legal landscape and institutional framework to address multilingual jailbreaks, while promoting linguistic inclusivity. These include mandatory assessments of multilingual capabilities and vulnerabilities, public opinion research, and state support for multilingual AI development. The measures aim to improve AI safety and functionality through EU policy initiatives, guiding the implementation of the EU AI Act and informing regulatory efforts of the European AI Office.

cs.CL↗

arXiv2024

Resolving the size and charge of small particles: a predictive model of nanopore mechanics

The movement of small particles and molecules through membranes is widespread and has far-reaching implications. Consequently, the development of mathematical models is essential for understanding these processes on a micro level, leading to deeper insights. In this endeavour, we suggested a model based on a set of empirical equations to predict the transport of substances through a solid-state nanopore and the associated signals generated during their translocation. This model establishes analytical relationships between the ionic current and electrical double-layer potential observed during ana-lyte translocation and their size, charge, and mobility in an electrolyte solution. This framework allows for rapid interpretation and prediction of the nanopore system's behaviour and provides a means for quantitatively determining the physical properties of molecular analytes. To illustrate the analyt-ical capability of this model, ceria nanoparticles were investigated while undergoing oxidation or reduction within an original nanopore device. The re-sults obtained were found to be in good agreement with predictions from physicochemical methods. This developed approach and model possess transfer-able utility to various porous materials, thereby expediting research efforts in membrane characterization and the advancement of nano- and ultrafiltra-tion or electrodialysis technologies.

physics.chem-ph↗

arXiv2024

Sparse Concept Bottleneck Models: Gumbel Tricks in Contrastive Learning

We propose a novel architecture and method of explainable classification with Concept Bottleneck Models (CBMs). While SOTA approaches to Image Classification task work as a black box, there is a growing demand for models that would provide interpreted results. Such a models often learn to predict the distribution over class labels using additional description of this target instances, called concepts. However, existing Bottleneck methods have a number of limitations: their accuracy is lower than that of a standard model and CBMs require an additional set of concepts to leverage. We provide a framework for creating Concept Bottleneck Model from pre-trained multi-modal encoder and new CLIP-like architectures. By introducing a new type of layers known as Concept Bottleneck Layers, we outline three methods for training them: with $\ell_1$-loss, contrastive loss and loss function based on Gumbel-Softmax distribution (Sparse-CBM), while final FC layer is still trained with Cross-Entropy. We show a significant increase in accuracy using sparse hidden layers in CLIP-based bottleneck models. Which means that sparse representation of concepts activation vector is meaningful in Concept Bottleneck Models. Moreover, with our Concept Matrix Search algorithm we can improve CLIP predictions on complex datasets without any additional training or fine-tuning. The code is available at: https://github.com/Andron00e/SparseCBM.

cs.CV↗

arXiv2023

Cross-Modal Conceptualization in Bottleneck Models

Concept Bottleneck Models (CBMs) assume that training examples (e.g., x-ray images) are annotated with high-level concepts (e.g., types of abnormalities), and perform classification by first predicting the concepts, followed by predicting the label relying on these concepts. The main difficulty in using CBMs comes from having to choose concepts that are predictive of the label and then having to label training examples with these concepts. In our approach, we adopt a more moderate assumption and instead use text descriptions (e.g., radiology reports), accompanying the images in training, to guide the induction of concepts. Our cross-modal approach treats concepts as discrete latent variables and promotes concepts that (1) are predictive of the label, and (2) can be predicted reliably from both the image and text. Through experiments conducted on datasets ranging from synthetic datasets (e.g., synthetic images with generated descriptions) to realistic medical imaging datasets, we demonstrate that cross-modal learning encourages the induction of interpretable concepts while also facilitating disentanglement. Our results also suggest that this guidance leads to increased robustness by suppressing the reliance on shortcut features.

cs.CL↗

arXiv2022

NEREL-BIO: A Dataset of Biomedical Abstracts Annotated with Nested Named Entities

This paper describes NEREL-BIO -- an annotation scheme and corpus of PubMed abstracts in Russian and smaller number of abstracts in English. NEREL-BIO extends the general domain dataset NEREL by introducing domain-specific entity types. NEREL-BIO annotation scheme covers both general and biomedical domains making it suitable for domain transfer experiments. NEREL-BIO provides annotation for nested named entities as an extension of the scheme employed for NEREL. Nested named entities may cross entity boundaries to connect to shorter entities nested within longer entities, making them harder to detect. NEREL-BIO contains annotations for 700+ Russian and 100+ English abstracts. All English PubMed annotations have corresponding Russian counterparts. Thus, NEREL-BIO comprises the following specific features: annotation of nested named entities, it can be used as a benchmark for cross-domain (NEREL -> NEREL-BIO) and cross-language (English -> Russian) transfer. We experiment with both transformer-based sequence models and machine reading comprehension (MRC) models and report their results. The dataset is freely available at https://github.com/nerel-ds/NEREL-BIO.

cs.CL↗

arXiv2022

Automatic generation of a large dictionary with concreteness/abstractness ratings based on a small human dictionary

Concrete/abstract words are used in a growing number of psychological and neurophysiological research. For a few languages, large dictionaries have been created manually. This is a very time-consuming and costly process. To generate large high-quality dictionaries of concrete/abstract words automatically one needs extrapolating the expert assessments obtained on smaller samples. The research question that arises is how small such samples should be to do a good enough extrapolation. In this paper, we present a method for automatic ranking concreteness of words and propose an approach to significantly decrease amount of expert assessment. The method has been evaluated on a large test set for English. The quality of the constructed dictionaries is comparable to the expert ones. The correlation between predicted and expert ratings is higher comparing to the state-of-the-art methods.

cs.CL↗

arXiv2022

RuNNE-2022 Shared Task: Recognizing Nested Named Entities

The RuNNE Shared Task approaches the problem of nested named entity recognition. The annotation schema is designed in such a way, that an entity may partially overlap or even be nested into another entity. This way, the named entity "The Yermolova Theatre" of type "organization" houses another entity "Yermolova" of type "person". We adopt the Russian NEREL dataset for the RuNNE Shared Task. NEREL comprises news texts written in the Russian language and collected from the Wikinews portal. The annotation schema includes 29 entity types. The nestedness of named entities in NEREL reaches up to six levels. The RuNNE Shared Task explores two setups. (i) In the general setup all entities occur more or less with the same frequency. (ii) In the few-shot setup the majority of entity types occur often in the training set. However, some of the entity types are have lower frequency, being thus challenging to recognize. In the test set the frequency of all entity types is even. This paper reports on the results of the RuNNE Shared Task. Overall the shared task has received 156 submissions from nine teams. Half of the submissions outperform a straightforward BERT-based baseline in both setups. This paper overviews the shared task setup and discusses the submitted systems, discovering meaning insights for the problem of nested NER. The links to the evaluation platform and the data from the shared task are available in our github repository: https://github.com/dialogue-evaluation/RuNNE.

cs.CL↗

arXiv2022

Extracting Software Requirements from Unstructured Documents

Requirements identification in textual documents or extraction is a tedious and error prone task that many researchers suggest automating. We manually annotated the PURE dataset and thus created a new one containing both requirements and non-requirements. Using this dataset, we fine-tuned the BERT model and compare the results with several baselines such as fastText and ELMo. In order to evaluate the model on semantically more complex documents we compare the PURE dataset results with experiments on Request For Information (RFI) documents. The RFIs often include software requirements, but in a less standardized way. The fine-tuned BERT showed promising results on PURE dataset on the binary sentence classification task. Comparing with previous and recent studies dealing with constrained inputs, our approach demonstrates high performance in terms of precision and recall metrics, while being agnostic to the unstructured textual input.

cs.SE↗

arXiv2021

NEREL: A Russian Dataset with Nested Named Entities, Relations and Events

In this paper, we present NEREL, a Russian dataset for named entity recognition and relation extraction. NEREL is significantly larger than existing Russian datasets: to date it contains 56K annotated named entities and 39K annotated relations. Its important difference from previous datasets is annotation of nested named entities, as well as relations within nested entities and at the discourse level. NEREL can facilitate development of novel models that can extract relations between nested named entities, as well as relations on both sentence and document levels. NEREL also contains the annotation of events involving named entities and their roles in the events. The NEREL collection is available via https://github.com/nerel-ds/NEREL.

cs.CL↗

arXiv2021

Surgical navigation systems based on augmented reality technologies

This study considers modern surgical navigation systems based on augmented reality technologies. Augmented reality glasses are used to construct holograms of the patient's organs from MRI and CT data, subsequently transmitted to the glasses. This, in addition to seeing the actual patient, the surgeon gains visualization inside the patient's body (bones, soft tissues, blood vessels, etc.). The solutions developed at Peter the Great St. Petersburg Polytechnic University allow reducing the invasiveness of the procedure and preserving healthy tissues. This also improves the navigation process, making it easier to estimate the location and size of the tumor to be removed. We describe the application of developed systems to different types of surgical operations (removal of a malignant brain tumor, removal of a cyst of the cervical spine). We consider the specifics of novel navigation systems designed for anesthesia, for endoscopic operations. Furthermore, we discuss the construction of novel visualization systems for ultrasound machines. Our findings indicate that the technologies proposed show potential for telemedicine.

cs.HC↗