arXiv subjects

Fedor Kovalev

Publications and source records attributed to Fedor Kovalev.

4 recordsLinked to original sources

arXiv2026

Amplified up-conversion of electromagnetic waves using time-varying metasurfaces

Time-varying metamaterials and photonic time crystals offer a powerful route to wave amplification through temporal modulation of material parameters. Here, we experimentally demonstrate amplified up-conversion of free-space electromagnetic waves in the microwave regime using a time-varying metasurface composed of varactor-loaded coupled split-ring resonators. The metasurface supports a cascaded parametric process in which temporal modulation enhances the resonant response near the signal frequency and subsequently enables efficient radiation at higher sum-frequency components. Direct measurements in degenerate and non-degenerate regimes show amplified up-conversion exceeding the single-step conversion estimate based on the Manley-Rowe frequency ratio. These results establish time-varying metasurfaces as an efficient platform for amplification, generation, and frequency conversion of electromagnetic waves in the microwave and terahertz bands, with potential extension to higher frequencies using ultrafast modulation mechanisms.

physics.optics↗

arXiv2025

Tunable bound states in the continuum through hybridization of 1D and 2D metasurfaces

This work presents a novel approach to create and dynamically control quasi-Bound States in the Continuum (BIC) resonances through the hybridization of 1D and 2D metasurfaces using micro-electromechanical systems (MEMS). By introducing out-of-plane symmetry breaking through a silicon MEMS membrane positioned above a 1D silicon metasurface, the quasi-BIC resonance's central wavelength and quality factor are precisely tuned. The proposed design achieves ultranarrow resonance linewidths with the spectral tuning range exceeding 60 nm while maintaining a constant quality factor. This tuning capability, realized through both horizontal displacement within a 1D metasurface and vertical MEMS membrane movement, offers a new degree of freedom for manipulating quasi-BIC resonances. The proposed hybridization of 2D and 1D metasurfaces using MEMS mechanism provides a practical route to dynamic modulation of transmission resonance characteristics, making it a promising candidate for tunable filters, spectroscopy, imaging, and sensing applications.

physics.optics↗

arXiv2021

Multi-Domain Adaptation in Neural Machine Translation Through Multidimensional Tagging

While NMT has achieved remarkable results in the last 5 years, production systems come with strict quality requirements in arbitrarily niche domains that are not always adequately covered by readily available parallel corpora. This is typically addressed by training domain specific models, using fine-tuning methods and some variation of back-translation on top of in-domain monolingual corpora. However, industrial practitioners can rarely afford to focus on a single domain. A far more typical scenario includes a set of closely related, yet succinctly different sub-domains. At Booking.com, we need to translate property descriptions, user reviews, as well as messages, (for example those sent between a customer and an agent or property manager). An editor might need to translate articles across a set of different topics. An e-commerce platform would typically need to translate both the description of each item and the user generated content related to them. To this end, we propose MDT: a novel method to simultaneously fine-tune on several sub-domains by passing multidimensional sentence-level information to the model during training and inference. We show that MDT achieves results competitive to N specialist models each fine-tuned on a single constituent domain, while effectively serving all N sub-domains, therefore cutting development and maintenance costs by the same factor. Besides BLEU (industry standard automatic evaluation metric known to only weakly correlate with human judgement) we also report rigorous human evaluation results for all models and sub-domains as well as specific examples that better contextualise the performance of each model in terms of adequacy and fluency. To facilitate further research, we plan to make the code available upon acceptance.

cs.CL↗

arXiv2017

Toward a full-scale neural machine translation in production: the Booking.com use case

While some remarkable progress has been made in neural machine translation (NMT) research, there have not been many reports on its development and evaluation in practice. This paper tries to fill this gap by presenting some of our findings from building an in-house travel domain NMT system in a large scale E-commerce setting. The three major topics that we cover are optimization and training (including different optimization strategies and corpus sizes), handling real-world content and evaluating results.

cs.CL↗