Search

Prior studies have shown that distinguishing text generated by large language models (LLMs) from human-written one is highly challenging, and often no better than random guessing. To verify the generalizability of this finding across languages and domains, we perform an extensive case study to identify the upper bound of human detection accuracy. Across 16 datasets covering 9 languages and 9 domains, 19 annotators achieved an average detection accuracy of 87.6%, thus challenging previous conclusions. We find that major gaps between human and machine text lie in concreteness, cultural nuances, and diversity. Prompting by explicitly explaining the distinctions in the prompts can partially bridge the gaps in over 50% of the cases. However, we also find that humans do not always prefer human-written text, particularly when they cannot clearly identify its source.

We report intense terahertz electroluminescence from Dirac-Landau polaritons, representing a major step toward achieving stimulated cyclotron emission and polariton-based lasers. By strongly coupling the cyclotron transitions of two-dimensional Dirac fermions in HgTe quantum wells with optical cavity modes, we observe efficient emission near the lasing threshold. This work demonstrates that polariton condensation, a process that bypasses the need for electronic population inversion, can significantly reduce the emission threshold compared to conventional mechanisms requiring high electric fields. Moreover, this concept unlocks the potential for stimulated emission in previously unsuitable narrow-gap semiconductors, as well as Dirac materials with non-equidistant Landau levels. These results open a new way for the development of compact, tunable terahertz lasers based on Landau polaritons, offering new opportunities for solid-state laser technology and applications in the terahertz gap., Comment: Manuscript and Supplementary Materials

In ``Characterization, stability and convergence of hierarchical clustering methods'' by G. E. Carlsson, F. Memoli, the natural way to construct an ultrametric space from a given metric space was presented. It was shown that the corresponding map $\textbf{U}$ is $1$-Lipschitz for every pair of bounded metric spaces, with respect to the Gromov-Hausdorff distance. We make a simple observation that $\textbf{U}$ is $1$-Lipschitz for pairs of all, not necessarily bounded, metric spaces. We then study the properties of the mapping $\textbf{U}$. We show that, for a given dotted connected metric space $A$, the mapping $\Psi\colon X\mapsto X\times A$ from the proper class of all bounded ultrametric spaces ($X\times A$ is endowed with the Manhattan metric) preserves the Gromov-Hausdorff distance. Moreover, the mapping $\textbf{U}$ is inverse to $\Psi$. By a dotted connected metric space, we mean a metric space in which for an arbitrary $\varepsilon > 0$ and every two points $p,\,q$, there exist points $x_0 = p,\,x_1,\,\ldots,\,x_n = q$ such that $\max_{0\le j \le n-1}|x_jx_{j+1}|\le \varepsilon$. At the end of the paper, we prove that each class (proper or not) consisting of unbounded metric spaces on finite Gromov-Hausdorff distances from each other cannot contain an ultrametric space and a dotted connected space simultaneously., Comment: 10 pages

We present a study of the radio and optical properties of the high-frequency peaker (HFP) blazar PKS 1614+051 at $z=3.21$ based on the data covering the time period of 1997-2024. The radio data are represented by the almost instantaneous 1-22 GHz measurements from the SAO RAS RATAN-600 radio telescope, the 5 and 8 GHz data from the IAA RAS RT-32 telescopes, and the 37 GHz data from the RT-22 telescope of CrAO RAS. The optical measurements in the $R$ band were collected with the SAO RAS 1-m Zeiss-1000 and 0.5-m AS-500/2 telescopes and the ZTF archive data. We have found low overall variability indices (0.1-0.2) and a spectral peak around 4.6 GHz, which is stable during the long-term period of monitoring. An analysis of the radio light curves reveals significant time delays (0.6 to 6.4 years) between the radio frequencies along with variability timescales ranging from 0.2 to 1.8 years in the source's rest frame, which is similar to the blazars at lower redshifts. Spectral modeling suggests the presence of both synchrotron-self absorption (SSA) and free-free absorption (FFA) processes. Based on the SSA model, we provide estimates of the magnetic field strength, which peaks at $\sim\!30$ mG. A spectroscopic study with the BTA SCORPIO-I spectrograph has found signs of the regular motion of a neutral hydrogen envelope around the blazar center, which confirms the presence of enough gaseous matter to form an external FFA screen. The results highlight the importance of multiwavelength and long-term monitoring to understand the physical mechanisms driving the variability in high-redshift blazars.

This paper introduces a new suite of question answering datasets for Norwegian; NorOpenBookQA, NorCommonSenseQA, NorTruthfulQA, and NRK-Quiz-QA. The data covers a wide range of skills and knowledge domains, including world knowledge, commonsense reasoning, truthfulness, and knowledge about Norway. Covering both of the written standards of Norwegian - Bokm{\aa}l and Nynorsk - our datasets comprise over 10k question-answer pairs, created by native speakers. We detail our dataset creation approach and present the results of evaluating 11 language models (LMs) in zero- and few-shot regimes. Most LMs perform better in Bokm{\aa}l than Nynorsk, struggle most with commonsense reasoning, and are often untruthful in generating answers to questions. All our datasets and annotation materials are publicly available., Comment: Accepted for NoDaLiDa / Baltic-HLT 2025

We present the GenAI Content Detection Task~1 -- a shared task on binary machine generated text detection, conducted as a part of the GenAI workshop at COLING 2025. The task consists of two subtasks: Monolingual (English) and Multilingual. The shared task attracted many participants: 36 teams made official submissions to the Monolingual subtask during the test phase and 26 teams -- to the Multilingual. We provide a comprehensive overview of the data, a summary of the results -- including system rankings and performance scores -- detailed descriptions of the participating systems, and an in-depth analysis of submissions. https://github.com/mbzuai-nlp/COLING-2025-Workshop-on-MGT-Detection-Task1, Comment: 18 pages

We introduce a dataset of high-quality human-authored summaries of news articles in Norwegian. The dataset is intended for benchmarking the abstractive summarisation capabilities of generative language models. Each document in the dataset is provided with three different candidate gold-standard summaries written by native Norwegian speakers, and all summaries are provided in both of the written variants of Norwegian -- Bokm{\aa}l and Nynorsk. The paper describes details on the data creation effort as well as an evaluation of existing open LLMs for Norwegian on the dataset. We also provide insights from a manual human evaluation, comparing human-authored to model-generated summaries. Our results indicate that the dataset provides a challenging LLM benchmark for Norwegian summarisation capabilities, Comment: Accepted at NoDaLiDa2025

In this paper, we study metric trees, without any finiteness restrictions. For subsets of such trees, a condition that guarantees that the Hausdorff and Gromov--Hausdorff distances from the subset to the entire metric tree are the same is obtained. This result allows to construct a new class of shortest geodesics (in the proper class of all metric spaces) connecting such subset of a metric tree with the tree itself. In particular, the technique elaborated is demonstrated on subsets of the real line., Comment: 10 pages, 0 figures

We give examples of cohomologies of the superconformal algebra, relevant to computations in the AdS supergravity. Our main examples are deformations of $AdS_5\times S^5$ transforming in finite-dimensional representations of the superconformal algebra at the linearized level. In the study of correlation functions, it is important to compute the resolution of BRST-exact products of vertex operators. The resolution is typically non-covariant, because of cohomological obstacles. Using the pure spinor formalism, we develop a framework to describe these obstacles, and formulate conjectures about their structure., Comment: 23 pp, LaTeX; v2: extended Introduction, added section on open questions

The ability to image electromagnetic fields holds key scientific and industrial applications, including electromagnetic compatibility, diagnostics of high-frequency devices, and experimental scientific work involving field interactions. Generally electric and magnetic field measurements require conductive elements which significantly distort the field. However, electromagnetic fields can be measured without altering the field via the shift they induce on Rydberg states of alkali atoms in atomic vapor, which are highly sensitive to electric fields. Previous field measurements using Rydberg atoms utilized electromagnetically induced transparency to read out the shift on the states induced by the fields, but did not provide spatial resolution. In this work, we demonstrate that electromagnetically induced transparency can be spatially resolved by imaging the fluorescence of the atoms. We demonstrate that this can be used to image $\sim$ V/cm scale electric fields in the DC-GHz range and $\sim$ mT scale static magnetic fields, with minimal distortion to the fields. We also demonstrate the ability to image $\sim$ 5 mV/cm scale fields for resonant microwave radiation and measure standing waves generated by the partial reflection of the vapor cell walls in this regime. With additional processing techniques like lock-in detection, we predict that our sensitivities could reach down to nV/cm levels. We perform this field imaging with a spatial resolution of 160 $\mu$m, limited by our imaging system, and estimate the fundamental resolution limitation to be 5 $\mu$m.

The use of copyrighted materials in training language models raises critical legal and ethical questions. This paper presents a framework for and the results of empirically assessing the impact of publisher-controlled copyrighted corpora on the performance of generative large language models (LLMs) for Norwegian. When evaluated on a diverse set of tasks, we found that adding both books and newspapers to the data mixture of LLMs tend to improve their performance, while the addition of fiction works seems to be detrimental. Our experiments could inform the creation of a compensation scheme for authors whose works contribute to AI development., Comment: 17 pages, 5 figures, 8 tables. Accepted at NoDaLiDa/Baltic-HLT 2025

Training large language models requires vast amounts of data, posing a challenge for less widely spoken languages like Norwegian and even more so for truly low-resource languages like Northern S\'ami. To address this issue, we present a novel three-stage continual training approach that substantially improves the downstream performance together with the inference efficiency for the target languages. Based on our findings, we train, evaluate, and openly release a new generative language model for Norwegian Bokm\r{a}l, Nynorsk, and Northern S\'ami with 11.4 billion parameters: NorMistral-11B., Comment: Published at NoDaLiDa 2025

We present a quantum optics-based detection method for determining the position and current of an electron beam. As electrons pass through a dilute vapor of rubidium atoms, their magnetic field perturb the atomic spin's quantum state and causes polarization rotation of a laser resonant with an optical transition of the atoms. By measuring the polarization rotation angle across the laser beam, we recreate a 2D projection of the magnetic field and use it to determine the e-beam position, size and total current. We tested this method for an e-beam with currents ranging from 30 to 110 {\mu}A. Our approach is insensitive to electron kinetic energy, and we confirmed that experimentally between 10 to 20 keV. This technique offers a unique platform for non-invasive characterization of charged particle beams used in accelerators for particle and nuclear physics research.

In this paper we prove that the Gromov--Hausdorff distance between $\mathbb{R}^n$ and its subset $A$ is finite if and only if $A$ is an $\varepsilon$-net in $\mathbb{R}^n$ for some $\varepsilon>0$. For infinite-dimensional Euclidean spaces this is not true. The proof is essentially based on upper estimate of the Euclidean Gromov--Hausdorff distance by means of the Gromov-Hausdorff distance., Comment: 6 pages, accepted for publication in Commun. Math. Res. (http://www.global-sci.org/intro/online.html?journal=cmr)

The rapid proliferation of large language models (LLMs) has increased the volume of machine-generated texts (MGTs) and blurred text authorship in various domains. However, most existing MGT benchmarks include single-author texts (human-written and machine-generated). This conventional design fails to capture more practical multi-author scenarios, where the user refines the LLM response for natural flow, coherence, and factual correctness. Our paper introduces the Benchmark of Expert-edited Machine-generated Outputs (Beemo), which includes 6.5k texts written by humans, generated by ten instruction-finetuned LLMs, and edited by experts for various use cases, ranging from creative writing to summarization. Beemo additionally comprises 13.1k machine-generated and LLM-edited texts, allowing for diverse MGT detection evaluation across various edit types. We document Beemo's creation protocol and present the results of benchmarking 33 configurations of MGT detectors in different experimental setups. We find that expert-based editing evades MGT detection, while LLM-edited texts are unlikely to be recognized as human-written. Beemo and all materials are publicly available., Comment: Accepted to NAACL 2025

We present a study of the multiwavelength (MW) variability of the blazar AO 0235+164 based on the radio-to-$\gamma$-ray data covering a long time period from 1997 to 2023. The radio data are represented by the 1-22 GHz measurements from the RATAN-600 radio telescope, the 5 and 8 GHz data from the RT-32 telescopes, and the 37 GHz data from the RT-22 telescope. The optical measurements in the $R$-band were collected with the 1-m Zeiss-1000 and 0.5-m AS-500/2 telescopes. Additionally we used the archive data at 230~GHz from the SMA and the $\gamma$-ray data in the 0.1-100 GeV band from the Fermi-LAT point source 4FGL-DR2 catalogue. The variability properties during four epochs containing major flares and one epoch of relatively low activity were analysed. A significant correlation ($\geq\!2\sigma$) between the radio, optical, and $\gamma$-ray bands is found for all these periods with time delays from 0 to 1.7 yrs. The relation between time delay and frequency is described by a linear law with a negative slope of -10 day/GHz. The discovered properties of MW variability for the low activity period and for flaring states suggest that the mechanisms dominating the radio-$\gamma$-ray variations are not substantially different. The detected quasi-periodic oscillations of about 6 and 2 years are tentative, as the time span of the observations includes fewer than 4 full cycles for the radio and optical data and only about 3 cycles for the Fermi-LAT data. The physical parameters of the radio jet were obtained using the Hedgehog model applied to the average radio spectrum of AO 0235+164 in the range 0.1-300 GHz. The effectiveness of replacing electrons with protons in the synchrotron radio emission of relativistic jets is shown for describing the nature of blazars and the generation of high energy neutrinos., Comment: Accepted 2024 October 30 for publication in MNRAS

This work sheds light on the electronic band properties of chiral WSSe Janus nanotubes from the quantum mechanical perspective. Line groups theory developed by Damnjanovich was used to model mechanical torsion of chiral nanotubes with different symmetries. Known natural torsion in chiral nanotubes was observed. It was shown that mechanical stress can be used as a tool to tune electronic properties of the nanotubes including the change of the nature of the electron transition. On the DFT-level of the theory the stability of WS2, WSe2 and Janus WSSe nanotubes was estimated. Applicability of stable Janus-nanotubes to photocatalytic water-splitting was suggested based on the calculated electronic properties, Comment: 9 pages, 9 figrues, 2 tables

We report the experimental observation of Anderson localization in two-dimensional (2D) electrons and holes in the bulk of HgTe quantum wells with a semimetallic spectrum and under strong disorder. In contrast, the one-dimensional (1D) edge channels, arising from the spectrum's inversion, demonstrate remarkable robustness against disorder due to topological protection. Strong disorder induces a mobility gap in the bulk, enabling access to the 1D edge states and thereby realizing the two-dimensional topological Anderson insulator (TAI) state. Nonlocal transport measurements confirm the emergence of topologically protected edge channels. The TAI state appears to be very sensitive to an external magnetic field applied perpendicular to the sample. Firstly, a small magnetic field of 30mT breaks the topological protection of 1D edge channels, thus turning the system into an ordinary Anderson insulator. Secondly, the magnetic field of 0.5T delocalizes 2D bulk electrons, transforming the system into a quantum Hall liquid.

We study the quantum Hall effect (QHE) in the three-dimensional topological insulator HgTe, which features topological Dirac-type surface states in a bulk gap opened by strain. Despite the co-existence of multiple carrier subsystems, the system exhibits perfectly quantized Hall plateaus at high magnetic fields. Here we study the system using three different experimental techniques: Transport experiments, capacitance measurements including the quantum capacitance, and current distribution measurements using electrostatically sensitive scanning probe microscopy. Our key finding is that at sufficiently high magnetic fields, the different electronic subsystems merge into one, and the current in a quantum Hall plateau is distributed across the entire width of the Hall bar device., Comment: 9 pages, 9 figures. Submitted to PRR

We explore electrical transport through a point contact in strained HgTe, a three-dimensional topological insulator. In the absence of a magnetic field $B$, there is no quantization. However, under higher magnetic fields, we observe distinct non-integer conductance steps. Based on numerical tight-binding calculations and a phenomenological Landauer-B\"uttiker approach, we attribute these atypical, non-integer quantized plateaus to significant scattering effects at the point contact., Comment: Submitted to PRR on 10/21/2024, re-submitted with minor changes to PRR on 01/14/2025, 9 pages, 7 figures

The multiplicities of positive and negative pions, kaons and unidentified hadrons produced in deep-inelastic scattering are measured in bins of the Bjorken scaling variable $x$, the relative virtual-photon energy $y$ and the fraction of the virtual-photon energy transferred to the final-state hadron $z$. Data were obtained by the COMPASS Collaboration using a 160 GeV muon beam of both electric charges and a liquid hydrogen target. These measurements cover the kinematic domain with photon virtuality $Q^2 > 1$ (GeV/$c)^2$, $0.004 < x < 0.4$, $0.1 < y < 0.7$ and $0.2 < z < 0.85$, in accordance with the kinematic domain used in earlier published COMPASS multiplicity measurements with an isoscalar target. The calculation of radiative corrections was improved by using the Monte Carlo generator DJANGOH, which results in up to 12\% larger corrections in the low-$x$ region., Comment: 19 pages, 29 figures

The quantum Hall effect, which exhibits a number of unusual properties, is studied in a gated 1000-nm-thick HgTe film, nominally a three-dimensional system. A weak zero plateau of Hall resistance, accompanied by a relatively small value of Rxx of the order of h/e^2, is found around the point of charge neutrality. It is shown that the zero plateau is formed by the counter-propagating chiral electron-hole edge channels, the scattering between which is suppressed. So, phenomenologically, the quantum spin Hall effect is reproduced, but with preserved ballisticity on macroscopic scales (larger than 1mm). It is shown that the formation of the QHE occurs in a two-dimensional (2D) accumulation layer near the gate, while the bulk carriers play the role of an electron reservoir. Due to the exchange of carriers between the reservoir and the 2D layer, an anomalous scaling of the QHE is observed not with respect to the CNP, but with respect to the first electron plateau., Comment: 12 pages, 10 figures

We report on the observation and comprehensive study of the terahertz radiation induced magneto-photogalvanic effect (MPGE) in bulk CdHgTe crystals hosting Kane fermions. The MPGE has been detected in Cd$_{x}$Hg$_{1-x}$Te films with Cd contents $x = 0.15$ and $0.22$ subjected to an in-plane magnetic field. At liquid helium temperature we observed multiple resonances in MPGE current upon variation of magnetic field. In the $x = 0.22$ with noninverted band structure, the resonances are caused by cyclotron resonance (CR) and photoionization of an impurity level. In the $x = 0.15$ films with an inverted band structure, they originate from the CR and interband optical transitions. Band structure calculated by the Kane model perfectly describes positions of all resonances. In particularly, the resonant MPGE caused by interband transitions excited by THz radiation is caused by the gapless energy spectrum of Kane fermions realized in materials with certain Cd contents and temperature range. In addition to the resonant MPGE current we detected a nonresonant one due to indirect optical transitions (Drude-like). This contribution has a nonmonotonic magnetic field dependence increasing linearly at low magnetic field $B$, approaching a maximum at moderate field and decreasing at high $B$. While the nonresonant MPGE decreases drastically with increasing temperature, it is well measurable up to room temperature. The developed theory demonstrates that the MPGE current arises due to cubic in momentum spin-dependent terms in the scattering probability. The asymmetry caused by these effects results in a pure spin current which is converted into an electric current due to the Zeeman effect.

The ease of access to large language models (LLMs) has enabled a widespread of machine-generated texts, and now it is often hard to tell whether a piece of text was human-written or machine-generated. This raises concerns about potential misuse, particularly within educational and academic domains. Thus, it is important to develop practical systems that can automate the process. Here, we present one such system, LLM-DetectAIve, designed for fine-grained detection. Unlike most previous work on machine-generated text detection, which focused on binary classification, LLM-DetectAIve supports four categories: (i) human-written, (ii) machine-generated, (iii) machine-written, then machine-humanized, and (iv) human-written, then machine-polished. Category (iii) aims to detect attempts to obfuscate the fact that a text was machine-generated, while category (iv) looks for cases where the LLM was used to polish a human-written text, which is typically acceptable in academic writing, but not in education. Our experiments show that LLM-DetectAIve can effectively identify the above four categories, which makes it a potentially useful tool in education, academia, and other domains. LLM-DetectAIve is publicly accessible at https://github.com/mbzuai-nlp/LLM-DetectAIve. The video describing our system is available at https://youtu.be/E8eT_bE7k8c.

This paper presents Papilusion, an AI-generated scientific text detector developed within the DAGPap24 shared task on detecting automatically generated scientific papers. We propose an ensemble-based approach and conduct ablation studies to analyze the effect of the detector configurations on the performance. Papilusion is ranked 6th on the leaderboard, and we improve our performance after the competition ended, achieving 99.46 (+9.63) of the F1-score on the official test set., Comment: to appear in "The 4th Workshop on Scholarly Document Processing @ ACL 2024" proceedings

We consider evolution (non-stationary) space-periodic solutions to the $n$-dimensional non-linear Navier-Stokes equations of anisotropic fluids with the viscosity coefficient tensor variable in space and time and satisfying the relaxed ellipticity condition. Employing the Galerkin algorithm, we prove the existence of Serrin-type solutions, that is, weak solutions with the velocity in the periodic space $L_2(0,T;\dot{\mathbf H}^{n/2}_{\#\sigma})$, $n\ge 2$. The solution uniqueness and regularity results are also discussed., Comment: 44 pages. arXiv admin note: text overlap with arXiv:2402.05792

In the present paper we study the original Gromov-Hausdorff distance between real normed spaces. In the first part of the paper we prove that two finite-dimensional real normed spaces on a finite Gromov-Hausdorff distance are isometric to each other. We then study the properties of finite point sets in finite-dimensional normed spaces whose cardinalities exceed the equilateral dimension of an ambient space. By means of the obtained results we prove the following enhancement of the aforementioned theorem: every finite-dimensional normed space lies on an infinite Gromov-Hausdorff distance from all other non-isometric normed spaces.

Minimal pairs are a well-established approach to evaluating the grammatical knowledge of language models. However, existing resources for minimal pairs address a limited number of languages and lack diversity of language-specific grammatical phenomena. This paper introduces the Russian Benchmark of Linguistic Minimal Pairs (RuBLiMP), which includes 45k pairs of sentences that differ in grammaticality and isolate a morphological, syntactic, or semantic phenomenon. In contrast to existing benchmarks of linguistic minimal pairs, RuBLiMP is created by applying linguistic perturbations to automatically annotated sentences from open text corpora and carefully curating test data. We describe the data collection protocol and present the results of evaluating 25 language models in various scenarios. We find that the widely used language models for Russian are sensitive to morphological and agreement-oriented contrasts but fall behind humans on phenomena requiring understanding of structural relations, negation, transitivity, and tense. RuBLiMP, the codebase, and other materials are publicly available., Comment: to appear in EMNLP 2024 (main)

Different ways exist to obtain the elements of the $\{\beta \}$-expansion for renormgroup invariant quantities. Here we consider independent confirmation within the standard QCD of a number of our results [1] for the values of elements of this expansion for the nonsinglet Adler $D_A$-function, Bjorken polarized sum rules $S^{Bjp}$ up to the order N$^4$LO. We suggest an approach to estimate the results of high order QCD calculations using a smaller number of diagrams of the specific type. This type is based on a proposed generalization of Naive NonAbelianization., Comment: In version 3:minor corrections to version 2, corresponds to journal publication JHEP10(2024)166. 14 pages,3 figures

We present a study of the radio variability of bright, $S_{1.4}\geq100$ mJy, high-redshift quasars at $z\geq3$ on timescales up to 30-40 years. The study involved simultaneous RATAN-600 measurements at frequencies of 2.3, 4.7, 8.2, 11.2, and 22.3 GHz in 2017-2020. In addition, data from the literature were used. We have found that the variability index, $V_S$, which quantifies the normalized difference between the maximum and minimum flux density while accounting for measurement uncertainties, ranges from 0.02 to 0.96 for the quasars. Approximately half of the objects in the sample exhibit a variability index within the range of 0.25 to 0.50, comparable to that observed in blazars at lower redshifts. The distribution of $V_S$ at 22.3 GHz is significantly different from that at 2.3-11.2 GHz, which may be attributed to the fact that a compact AGN core dominates at the source's rest frame frequencies greater than 45 GHz, leading to higher variability indices obtained at 22.3 GHz (the $V_S$ distribution peaks around 0.4) compared to the lower frequencies (the $V_S$ distribution at 2.3 and 4.7 GHz peaks around 0.1-0.2). Several source groups with distinctive variability characteristics were found using cluster analysis of quasars. We propose 7 new candidates for gigahertz peaked-spectrum (GPS) sources and 5 new megahertz peaked-spectrum (MPS) sources based on their spectrum shape and variability features. Only 6 out of 23 sources previously reported as GPS demonstrate a low variability level typical of classical GPS sources ($V_{S} < 0.25$) at 4.7-22.3 GHz. When excluding the highly variable peaked-spectrum blazars, we expect no more than 20% of the sources in the sample to be GPS candidates and no more than 10% to be MPS candidates., Comment: 56 pages, 20 Figures. Published in MDPI Galaxies

Shubnikov-de Haas (SdH) oscillations and magneto-intersubband oscillations of magnetoresistance of structures with single HgTe quantum wells with a width of (10-18) nm have been experimentally studied. The spectrum of the conduction band in these structures is split by the spin-orbit interaction. This leads to beats of the SdH oscillations and the appearance of low-frequency magneto-intersubband oscillations. The mutual position of the antinodes of the SdH oscillations and the maxima of the magneto-interband oscillations is unusual -- in low magnetic fields it is directly opposite to the predictions of the theory. Measurements in high magnetic fields, in which the relative amplitude of the SdH oscillations becomes greater than 0.2-0.3, show a change in the relative position of the antinodes of the SdH oscillations and the maxima of low-frequency oscillations. Numerical calculations and additional measurements at different temperatures show that the observed effects are due to oscillations of the Fermi level in the magnetic field., Comment: 7 two-column pages, 10 figures

The approach to the theory of a relativistic random process is considered by the path integral method as Brownian motion taking into account the boundedness of speed. An attempt was made to build a relativistic analogue of the Wiener measure as a weak limit of finite-difference approximations. A formula has been proposed for calculating the probability particle transition during relativistic Brownian motion. Calculations were carried out by three different methods with identical results. Along the way, exact and asymptotic formulas for the volume of some parts and sections of an N-1-dimensional unit cube were obtained. They can have independent value.

In this letter, we revisit the quantisation problem for a fundamental model of classical mechanics - the Zhukovsky-Volterra top. We have discovered a four-parametric pencil of compatible Poisson brackets, comprising two quadratic and two linear Poisson brackets. Using the quantisation ideal method, we have identified two distinct quantisations of the Zhukovsky-Volterra top. The first type corresponds to the universal enveloping algebras of $so(3)$, leading to Lie-Poisson brackets in the classical limit. The second type can be regarded as a quantisation of the four-parametric inhomogeneous quadratic Poisson pencil. We discuss the relationships between the quantisations obtained in our paper, Sklyanin's quantisation of the Euler top, and Levin-Olshanetsky-Zotov's quantisation of the Zhukovsky-Volterra top.

In this paper, we study operations on functors in the category of abelian groups simplar to the derivation in the sense of Dold-Puppe. They are defined as derived limits of a functor applied to the relation subgroup over a category of free presentations of the group. The integral homology of the Eilenberg-Maclane space $K(\mathbb Z,3)$ appears as a part of description of these operations applied to symmetric powers.

For a functor from the category of free presentations of a group to the category of all groups we define the boundary limit as an image of the natural map from limit to colimit. We show that the fourth dimension quotient of a group can be naturally described as the boundary limit of a simply-defined functor.

The resistivity of two-dimensional (2D) metals generally exhibits insensitivity to electron-electron scattering. However, it's worth noting that Galilean invariance may not hold true in systems characterized by a spectrum containing multiple electronic branches or in scenarios involving electron-hole plasma. In the context of our study, we focus on 2D electrons confined within a triple quantum well (TQW) based on HgTe. This system displays a coexistence of energy bands featuring both linear and parabolic-like spectra at low energy and, therefore, lacks the Galilean invariance. This research employs a combined theoretical and experimental approach to investigate the transport properties of this two-component system across various regimes. By manipulating carrier density and temperature, we tune our system from a fully degenerate regime, where resistance follows a temperature-dependent behavior proportional to $T^2$, to a regime where both types of electrons adhere to Boltzmann statistics. In the non-degenerate regime, electron interactions lead to resistance that is weakly dependent on temperature. Notably, our experimental observations closely align with the theoretical predictions derived in this study. This work establishes the HgTe-based TQW as a promising platform for exploring different interaction dominant scenarios for the massless-massive Dirac system.9 pages, 8 figures, Comment: 12 pages, 8 figures

Materials with linear electronic dispersion often feature high carrier mobilities and unusually strong nonlinear optical interactions. In this work, we investigate the THz nonlinear dynamics of one such material, HgCdTe, with an electronic band dispersion heavily dependent on both temperature and stoichiometry. We show how the band gap, carrier concentration and band shape together determine the nonlinear response of the system. At low temperatures, carrier generation from Zener tunneling dominates the nonlinear response with a reduction in the overall transmission. At room temperature, quasi-ballistic electronic dynamics drive the largest observed nonlinear optical interactions, leading to a transmission increase. Our results demonstrate the sensitivity of these nonlinear optical properties of narrow-gap materials to small changes in the electronic dispersion and carrier concentration., Comment: 48 pages, 21 figures

Based on the collected multiwavelength data, namely in the radio (NVSS, FIRST, RATAN-600), IR (WISE), optical (Pan-STARRS), UV (GALEX), and X-ray (ROSAT, Swift-XRT) ranges, we have performed a cluster analysis for the blazars of the Roma-BZCAT catalog. Using two machine learning methods, namely a combination of PCA with k-means clustering and Kohonen's self-organizing maps, we have constructed an independent classification of the blazars (five classes) and compared the classes with the known Roma-BZCAT classification (FSRQs, BL Lacs, galaxy-dominated BL Lacs, and blazars of an uncertain type) as well as with the high synchrotron peaked blazars (HSP) from the 3HSP catalog and blazars from the TeVCat catalog. The obtained groups demonstrate concordance with the BL Lac/FSRQ classification along with a continuous character of the change in the properties. The group of HSP blazars stands out against the overall distribution. We examine the characteristics of the five groups and demonstrate distinctions in their spectral energy distribution shapes. The effectiveness of the clustering technique for objective analysis of multiparametric arrays of experimental data is demonstrated., Comment: Accepted for publication in Research in Astronomy and Astrophysics

This paper describes AIpom, a system designed to detect a boundary between human-written and machine-generated text (SemEval-2024 Task 8, Subtask C: Human-Machine Mixed Text Detection). We propose a two-stage pipeline combining predictions from an instruction-tuned decoder-only model and encoder-only sequence taggers. AIpom is ranked second on the leaderboard while achieving a Mean Absolute Error of 15.94. Ablation studies confirm the benefits of pipelining encoder and decoder models, particularly in terms of improved performance., Comment: 2nd place at SemEval-2024 Task 8, Subtask C, to appear in SemEval-2024 proceedings

The semimetal quantum well (QW) based on HgTe structures exhibiting unusual transport properties at low temperature is examined experimentally. It demonstrates either a linear or quadratic growth of resistance with temperature at different top-gate voltages in the semimetal regime. We develop a theoretical model of HgTe-based semimetal QW resistance temperature dependence based on electron-hole scattering processes at low temperatures. We apply the Boltzmann transport equation approach to study the effect of electron-hole scattering in a semimetal QW. The calculated temperature behavior of 2D semimetal resistivity demonstrates an excellent agreement with experimental findings.