Oleg Yurievich Rogov
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Oleg Yurievich Rogov has a total of 31 co-authors. Here are the top co-authors by number of shared publications:
- Dmitry V. Dylov
Artificial Intelligence Research Institute, Moscow, Russian Federation and Skolkovo Institute of Science and Technology, Moscow, Russian Federation
4 shared publications - Alexander A. Ezhov
Faculty of Physics, Lomonosov Moscow State University, Moscow, Russia
2 shared publications - Darya Valeryaevna Pugacheva
HSE University, Moscow, Russian Federation and Domain-specific NLP Group, Moscow, Russian Federation
2 shared publications - Dmitrii Evgenievich Indenbom2 shared publications
- Dmitrii Sergeevich Korzh2 shared publications
- Elena Tutubalina
Kazan Federal University and HSE University, Moscow, Russia
2 shared publications
Latest Publications
- Open Access
- 4 December 2025
- Published by Kazan Federal University in Russian Digital Libraries Journal
- Vol. 28 (5)
Hiding in Meaning: Semantic Encoding for Generative Text Steganography
We propose a novel framework for steganographic text generation that hides binary messages within semantically coherent natural language using latent-space conditioning of large language models (LLMs). Secret messages are first encoded into continuous vectors via a learned binary-to-latent mapping, which is used to guide text generation through prefix tuning. Unlike prior token-level or syntactic steganography, our method avoids explicit word manipulation and instead operates entirely within the latent semantic space, enabling more fluent and less detectable outputs. On the receiver side, the latent representation is recovered from the generated text and decoded back into the original message. As a key theoretical contribution, we provide a robustness guarantee: if the recovered latent vector lies within a bounded distance of the original, exact message reconstruction is ensured, with the bound determined by the decoder’s Lipschitz continuity and the minimum logit margin. This formal result offers a principled view of the reliability–capacity trade-off in latent steganographic systems. Empirical evaluation on both synthetic data and real-world domains such as Amazon reviews shows that our method achieves high message recovery accuracy (above 91%), strong text fluency and competitive capacity up to 6 bits per sentence element while maintaining resilience against neural steganalysis. These findings demonstrate that latent conditioned generation offers a secure and practical pathway for embedding information in modern LLMs.
- 4 December 2025
- Published by Kazan Federal University in Russian Digital Libraries Journal
- Vol. 28 (5)
Open AccessHiding in Meaning: Semantic Encoding for Generative Text Steganography
We propose a novel framework for steganographic text generation that hides binary messages within semantically coherent natural language using latent-space conditioning of large language models (LLMs). Secret messages are first encoded into continuous vectors via a learned binary-to-latent mapping, which is used to guide text generation through prefix tuning. Unlike prior token-level or syntactic steganography, our method avoids explicit word manipulation and instead operates entirely within the latent semantic space, enabling more fluent and less detectable outputs. On the receiver side, the latent representation is recovered from the generated text and decoded back into the original message. As a key theoretical contribution, we provide a robustness guarantee: if the recovered latent vector lies within a bounded distance of the original, exact message reconstruction is ensured, with the bound determined by the decoder’s Lipschitz continuity and the minimum logit margin. This formal result offers a principled view of the reliability–capacity trade-off in latent steganographic systems. Empirical evaluation on both synthetic data and real-world domains such as Amazon reviews shows that our method achieves high message recovery accuracy (above 91%), strong text fluency and competitive capacity up to 6 bits per sentence element while maintaining resilience against neural steganalysis. These findings demonstrate that latent conditioned generation offers a secure and practical pathway for embedding information in modern LLMs.
- 1 December 2025
- Published by Allerton Press in Automatic Documentation and Mathematical Linguistics
- Vol. 59 (S5) , S447-S452
Hiding in Meaning: Semantic Encoding for Generative Text Steganography
- 1 December 2025
- Published by Allerton Press in Automatic Documentation and Mathematical Linguistics
- Vol. 59 (S5) , S447-S452
Hiding in Meaning: Semantic Encoding for Generative Text Steganography
- 10 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 3893-3906
GLiRA: Closed-Box Membership Inference Attack via Knowledge Distillation
While Deep Neural Networks demonstrate remarkable performance in practical tasks, they are vulnerable to membership inference attacks aimed at identifying whether a certain object belongs to the training dataset. To conduct a membership inference attack on a target model, an adversary has to train a set of shadow models and conduct a statistical test to determine the membership status of the particular input object. Usually, shadow models are trained without taking into account the target model; we argue that utilizing the predictions of the target model can guide the training process of the shadow model. To improve the efficiency of shadow model-based membership inference attacks, we propose GLiRA, a knowledge distillation-guided approach to membership inference attacks. We observe that the knowledge distillation significantly improves the efficiency of a likelihood ratio membership inference attack when the architecture of the target model is both known and unknown to an attacker. We evaluate the proposed method across multiple image classification datasets and model architectures and demonstrate that knowledge distillation-guided likelihood ratio attack outperforms the current state-of-the-art membership inference attacks in the majority of experimental settings.
- 10 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 20 (15566013) , 3893-3906
GLiRA: Closed-Box Membership Inference Attack via Knowledge Distillation
While Deep Neural Networks demonstrate remarkable performance in practical tasks, they are vulnerable to membership inference attacks aimed at identifying whether a certain object belongs to the training dataset. To conduct a membership inference attack on a target model, an adversary has to train a set of shadow models and conduct a statistical test to determine the membership status of the particular input object. Usually, shadow models are trained without taking into account the target model; we argue that utilizing the predictions of the target model can guide the training process of the shadow model. To improve the efficiency of shadow model-based membership inference attacks, we propose GLiRA, a knowledge distillation-guided approach to membership inference attacks. We observe that the knowledge distillation significantly improves the efficiency of a likelihood ratio membership inference attack when the architecture of the target model is both known and unknown to an attacker. We evaluate the proposed method across multiple image classification datasets and model architectures and demonstrate that knowledge distillation-guided likelihood ratio attack outperforms the current state-of-the-art membership inference attacks in the majority of experimental settings.
- 26 January 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 189, 188-194
FS-Net: Full scale network and adaptive threshold for improving extraction of micro-retinal vessel structures
- 26 January 2025
- Published by Elsevier in Pattern Recognition Letters
- Vol. 189, 188-194
FS-Net: Full scale network and adaptive threshold for improving extraction of micro-retinal vessel structures
- Open Access
- 9 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 85214-85223
Detect to Focus: Latent-Space Autofocusing System With Decentralized Hierarchical Multi-Agent Reinforcement Learning
State-of-the-art object detection models are frequently trained offline using available datasets, such as ImageNet: large and overly diverse data that are unbalanced and hard to cluster semantically. This kind of training drops the object detection performance should the change in illumination, in the environmental conditions (e.g., rain or dust), or in the lens positioning (out-of-focus blur) occur. We propose a simple way to intelligently control the camera and the lens focusing settings in such scenarios using DASHA, a Decentralized Autofocusing System with Hierarchical Agents. Our agents learn to focus on scenes in challenging environments, significantly enhancing the pattern recognition capacity beyond the popular detection models (YOLO, Faster R-CNN, and Retina are considered). At the same time, the decentralized training allows preserving the equipment from overheating. The algorithm relies on the latent representation of the camera’s stream and, thus, it is the first method to allow a completely no-reference imaging, where the system trains itself to auto-focus itself.
- 9 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 85214-85223
Open AccessDetect to Focus: Latent-Space Autofocusing System With Decentralized Hierarchical Multi-Agent Reinforcement Learning
State-of-the-art object detection models are frequently trained offline using available datasets, such as ImageNet: large and overly diverse data that are unbalanced and hard to cluster semantically. This kind of training drops the object detection performance should the change in illumination, in the environmental conditions (e.g., rain or dust), or in the lens positioning (out-of-focus blur) occur. We propose a simple way to intelligently control the camera and the lens focusing settings in such scenarios using DASHA, a Decentralized Autofocusing System with Hierarchical Agents. Our agents learn to focus on scenes in challenging environments, significantly enhancing the pattern recognition capacity beyond the popular detection models (YOLO, Faster R-CNN, and Retina are considered). At the same time, the decentralized training allows preserving the equipment from overheating. The algorithm relies on the latent representation of the camera’s stream and, thus, it is the first method to allow a completely no-reference imaging, where the system trains itself to auto-focus itself.
- Open Access
Medical Image Captioning via Generative Pretrained Transformers
The automatic clinical caption generation problem is referred to as proposed model combining the analysis of frontal chest X-Ray scans with structured patient information from the radiology records. We combine two language models, the Show-Attend-Tell and the GPT-3, to generate comprehensive and descriptive radiology records. The proposed combination of these models generates a textual summary with the essential information about pathologies found, their location, and the 2D heatmaps localizing each pathology on the original X-Ray scans. The proposed model is tested on two medical datasets, the Open-I, MIMIC-CXR, and the general-purpose MS-COCO. The results measured with the natural language assessment metrics prove their efficient applicability to the chest X-Ray image captioning.
Open AccessMedical Image Captioning via Generative Pretrained Transformers
The automatic clinical caption generation problem is referred to as proposed model combining the analysis of frontal chest X-Ray scans with structured patient information from the radiology records. We combine two language models, the Show-Attend-Tell and the GPT-3, to generate comprehensive and descriptive radiology records. The proposed combination of these models generates a textual summary with the essential information about pathologies found, their location, and the 2D heatmaps localizing each pathology on the original X-Ray scans. The proposed model is tested on two medical datasets, the Open-I, MIMIC-CXR, and the general-purpose MS-COCO. The results measured with the natural language assessment metrics prove their efficient applicability to the chest X-Ray image captioning.
- 8 November 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 6, 1586-1591
Adaptive Denoising and Alignment Agents for Infrared Imaging
Clinical infrared imagers, such as those used to visualize subcutaneous vasculature, rely on the image-projection feedback loops. Specifically, these instruments embed model-based feedback algorithms to process the acquired ’invisible’ infrared data streams and then to project a visible copy of them back to the imaged region (e.g., the skin of the patient’s forearm). Being inherently noisy, the infrared frames and their projections are prone to misalignment, demanding frequent instrument tuning and recalibration. To address this challenge, we propose to reconsider the feedback loop entailed in such imagers from the standpoint of multi-agent deep learning. Both proposed agents – Denoiser and Aligner – are adaptive in that they continuously optimize the corresponding target value functions. Namely, the Denoiser learns the proper frequency decomposition of the acquired infrared data until the target segmentation metric is maximized; whereas, the Aligner learns the intensity fluctuations within the segmentation mask tuned by the Denoiser until its maximal overlap with the source infrared image. The idea is validated synthetically on a benchmark dataset and experimentally on a bench-top vein scanner in the lab, with the duet of agents proving efficient in handling the distortions and the misalignment.
- 8 November 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 6, 1586-1591
Adaptive Denoising and Alignment Agents for Infrared Imaging
Clinical infrared imagers, such as those used to visualize subcutaneous vasculature, rely on the image-projection feedback loops. Specifically, these instruments embed model-based feedback algorithms to process the acquired ’invisible’ infrared data streams and then to project a visible copy of them back to the imaged region (e.g., the skin of the patient’s forearm). Being inherently noisy, the infrared frames and their projections are prone to misalignment, demanding frequent instrument tuning and recalibration. To address this challenge, we propose to reconsider the feedback loop entailed in such imagers from the standpoint of multi-agent deep learning. Both proposed agents – Denoiser and Aligner – are adaptive in that they continuously optimize the corresponding target value functions. Namely, the Denoiser learns the proper frequency decomposition of the acquired infrared data until the target segmentation metric is maximized; whereas, the Aligner learns the intensity fluctuations within the segmentation mask tuned by the Denoiser until its maximal overlap with the source infrared image. The idea is validated synthetically on a benchmark dataset and experimentally on a bench-top vein scanner in the lab, with the duet of agents proving efficient in handling the distortions and the misalignment.
- Open Access
- 3 October 2018
- Published by Financial University under the Government of the Russian Federation in Finance: Theory and Practice
- Vol. 22 (4) , 104-113
THE INFLUENCE OF THE TONALITY OF NEWS ON THE EXCHANGE RATE OF BITCOIN
The authors assess the impact of the emotional tonality of bitcoin news on its exchange rate. In particular, we studied the hypothesis of the impact of the readability index of the news text on the volatility of bitcoin. Despite the fact that excessive volatility threatens bitcoin not to become a successful currency, many scientists are interested in the determinants of such volatility. Factors such as speculative investments or the attention of the society are the drivers of the volatility of the exchange rate of bitcoin. In this regard, the question of studying the impact of news on the bitcoin exchange rate is relevant. The purpose of this paper is to assess the impact of the emotional tonality of bitcoin news on its exchange rate. The empirical base of the study was quite extensive since it includes more than 1330 news from the Thomson Reuters information base for the period from 19.08.2011 to 16.08.2016 on the bitcoin market. The research methodology includes the sentiment analysis conducted by using the dictionary MacDonald and Loughran and also the analysis of the interdependence of time series-based causal analysis using the test of Granger causation. We present three hypotheses about the impact of news on the bitcoin exchange rate. During the study, two of them were confirmed. We proved the first hypothesis that the negative news had a more significant impact than positive ones, taking into account the five time-lags. The second hypothesis about the impact of positive tonality in the news on the bitcoin exchange rate, using the Granger test for causation, was not confirmed, since the positive values of this test were obtained in two time-lags out of five. We can confirm that the third hypothesis was proved — the high readability index has an impact on the bitcoin volatility for the entire studied period, taking into account all five time-lags. Thus, the assumption about the impact of the emotional tonality of news on the bitcoin exchange rate can be confirmed.
- 3 October 2018
- Published by Financial University under the Government of the Russian Federation in Finance: Theory and Practice
- Vol. 22 (4) , 104-113
Open AccessTHE INFLUENCE OF THE TONALITY OF NEWS ON THE EXCHANGE RATE OF BITCOIN
The authors assess the impact of the emotional tonality of bitcoin news on its exchange rate. In particular, we studied the hypothesis of the impact of the readability index of the news text on the volatility of bitcoin. Despite the fact that excessive volatility threatens bitcoin not to become a successful currency, many scientists are interested in the determinants of such volatility. Factors such as speculative investments or the attention of the society are the drivers of the volatility of the exchange rate of bitcoin. In this regard, the question of studying the impact of news on the bitcoin exchange rate is relevant. The purpose of this paper is to assess the impact of the emotional tonality of bitcoin news on its exchange rate. The empirical base of the study was quite extensive since it includes more than 1330 news from the Thomson Reuters information base for the period from 19.08.2011 to 16.08.2016 on the bitcoin market. The research methodology includes the sentiment analysis conducted by using the dictionary MacDonald and Loughran and also the analysis of the interdependence of time series-based causal analysis using the test of Granger causation. We present three hypotheses about the impact of news on the bitcoin exchange rate. During the study, two of them were confirmed. We proved the first hypothesis that the negative news had a more significant impact than positive ones, taking into account the five time-lags. The second hypothesis about the impact of positive tonality in the news on the bitcoin exchange rate, using the Granger test for causation, was not confirmed, since the positive values of this test were obtained in two time-lags out of five. We can confirm that the third hypothesis was proved — the high readability index has an impact on the bitcoin volatility for the entire studied period, taking into account all five time-lags. Thus, the assumption about the impact of the emotional tonality of news on the bitcoin exchange rate can be confirmed.
- Open Access
- 2 August 2018
- Published by Springer Nature in Scientific Reports
- Vol. 8 (1) , 11623
Chiral visible light metasurface patterned in monocrystalline silicon by focused ion beam
High refractive index makes silicon the optimal platform for dielectric metasurfaces capable of versatile control of light. Among various silicon modifications, its monocrystalline form has the weakest visible light absorption but requires a careful choice of the fabrication technique to avoid damage, contamination or amorphization. Presently prevailing chemical etching can shape thin silicon layers into two-dimensional patterns consisting of strips and posts with vertical walls and equal height. Here, the possibility to create silicon nanostructure of truly tree-dimensional shape by means of the focused ion beam lithography is explored, and a 300 nm thin film of monocrystalline epitaxial silicon on sapphire is patterned with a chiral nanoscale relief. It is demonstrated that exposing silicon to the ion beam causes a substantial drop of the visible transparency, which, however, is completely restored by annealing with oxidation of the damaged surface layer. As a result, the fabricated chiral metasurface combines high (50–80%) transmittance with the circular dichroism of up to 0.5 and the optical activity of up to 20° in the visible range. Being also remarkably durable, it possesses crystal-grade hardness, heat resistance up to 1000 °C and the inertness of glass.
- 2 August 2018
- Published by Springer Nature in Scientific Reports
- Vol. 8 (1) , 11623
Open AccessChiral visible light metasurface patterned in monocrystalline silicon by focused ion beam
High refractive index makes silicon the optimal platform for dielectric metasurfaces capable of versatile control of light. Among various silicon modifications, its monocrystalline form has the weakest visible light absorption but requires a careful choice of the fabrication technique to avoid damage, contamination or amorphization. Presently prevailing chemical etching can shape thin silicon layers into two-dimensional patterns consisting of strips and posts with vertical walls and equal height. Here, the possibility to create silicon nanostructure of truly tree-dimensional shape by means of the focused ion beam lithography is explored, and a 300 nm thin film of monocrystalline epitaxial silicon on sapphire is patterned with a chiral nanoscale relief. It is demonstrated that exposing silicon to the ion beam causes a substantial drop of the visible transparency, which, however, is completely restored by annealing with oxidation of the damaged surface layer. As a result, the fabricated chiral metasurface combines high (50–80%) transmittance with the circular dichroism of up to 0.5 and the optical activity of up to 20° in the visible range. Being also remarkably durable, it possesses crystal-grade hardness, heat resistance up to 1000 °C and the inertness of glass.
- 3 October 2017
- Published by Wiley in Journal of Microscopy
- Vol. 268 (3) , 254-258
FIB‐fabricated complex‐shaped 3D chiral photonic silicon nanostructures
Summary: Technologies capable of fabricating complex shaped silicon metasurfaces attract increasing attention. The focused ion beam fabrication technique is considered traditionally as causing thick damaged layers in silicon resulting in a significant rise of the optical absorption loss. We examine the structure of the FIB‐fabricated nanostructures on the silicon‐on‐sapphire (SOS) platform and its optical characteristics before and after thermal oxidation. We show that being thermally oxidised the FIB‐patterned silicon subwavelength nanostructure tends to regain its chiral optical features. The impact of the oxidation process on the silicon nanostructure optical behaviour is discussed.
- 3 October 2017
- Published by Wiley in Journal of Microscopy
- Vol. 268 (3) , 254-258
FIB‐fabricated complex‐shaped 3D chiral photonic silicon nanostructures
Summary: Technologies capable of fabricating complex shaped silicon metasurfaces attract increasing attention. The focused ion beam fabrication technique is considered traditionally as causing thick damaged layers in silicon resulting in a significant rise of the optical absorption loss. We examine the structure of the FIB‐fabricated nanostructures on the silicon‐on‐sapphire (SOS) platform and its optical characteristics before and after thermal oxidation. We show that being thermally oxidised the FIB‐patterned silicon subwavelength nanostructure tends to regain its chiral optical features. The impact of the oxidation process on the silicon nanostructure optical behaviour is discussed.