-
Arakelov geometry of Cuntz-Pimsner algebras
Authors:
Igor V. Nikolaev
Abstract:
We use $K$-theory of the $C^*$-algebras to study the Arakelov geometry, i.e. a compactification of the arithmetic schemes $V\to Spec ~\mathbf{Z}$. In particular, it is proved that the Picard group of $V$ is isomorphic to the $K_0$-group of a Cuntz-Pimsner algebra associated to $V$. We apply the result to the finiteness problem for the algebraic varieties over number fields.
We use $K$-theory of the $C^*$-algebras to study the Arakelov geometry, i.e. a compactification of the arithmetic schemes $V\to Spec ~\mathbf{Z}$. In particular, it is proved that the Picard group of $V$ is isomorphic to the $K_0$-group of a Cuntz-Pimsner algebra associated to $V$. We apply the result to the finiteness problem for the algebraic varieties over number fields.
△ Less
Submitted 24 June, 2024;
originally announced June 2024.
-
Multimodal reconstruction of TbCo thin film structure with Basyeian analysis of polarised neutron reflectivity
Authors:
P. S. Savchenkov,
K. V. Nikolaev,
V. I. Bodnarchuk,
A. N. Pirogov,
A. V. Belushkin,
S. N. Yakunin
Abstract:
We implemented the Bayesian analysis to the polarised neutron reflectivity data. Reflectivity data from a magnetic TbCo thin film structure was studied using the bundle of a Monte-Carlo Markov-chain algorithm, likelihood estimation, and error modeling. By utilizing the Bayesian analysis, we were able to investigate the uniqueness of the solution beyond reconstructing the magnetic and structure par…
▽ More
We implemented the Bayesian analysis to the polarised neutron reflectivity data. Reflectivity data from a magnetic TbCo thin film structure was studied using the bundle of a Monte-Carlo Markov-chain algorithm, likelihood estimation, and error modeling. By utilizing the Bayesian analysis, we were able to investigate the uniqueness of the solution beyond reconstructing the magnetic and structure parameters. This approach has demonstrated its expedience as several probable reconstructions were found (the multimodality case) concerning the isotopic composition of the surface cover layer. Such multimodal reconstruction emphasizes the importance of rigorous data analysis instead of the direct data fitting approach, especially in the case of poor statistically conditioned data, typical for neutron reflectivity experiments. The analysis details and the discussion on multimodality are in this article.
△ Less
Submitted 2 May, 2024;
originally announced May 2024.
-
Local factors and Cuntz-Pimsner algebras
Authors:
Igor V. Nikolaev
Abstract:
We recast the local factors of the Hasse-Weil zeta function at infinity in terms of the Cuntz-Pimsner algebras. The nature of such factors is an open problem studied by Deninger and Serre.
We recast the local factors of the Hasse-Weil zeta function at infinity in terms of the Cuntz-Pimsner algebras. The nature of such factors is an open problem studied by Deninger and Serre.
△ Less
Submitted 18 April, 2024;
originally announced April 2024.
-
Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
Authors:
Gemini Team,
Petko Georgiev,
Ving Ian Lei,
Ryan Burnell,
Libin Bai,
Anmol Gulati,
Garrett Tanzer,
Damien Vincent,
Zhufeng Pan,
Shibo Wang,
Soroosh Mariooryad,
Yifan Ding,
Xinyang Geng,
Fred Alcober,
Roy Frostig,
Mark Omernick,
Lexi Walker,
Cosmin Paduraru,
Christina Sorokin,
Andrea Tacchetti,
Colin Gaffney,
Samira Daruki,
Olcan Sercinoglu,
Zach Gleicher,
Juliette Love
, et al. (1092 additional authors not shown)
Abstract:
In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February…
▽ More
In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.
△ Less
Submitted 14 June, 2024; v1 submitted 8 March, 2024;
originally announced March 2024.
-
On 1-skeleton of the cut polytopes
Authors:
Andrei V. Nikolaev
Abstract:
Given an undirected graph $G = (V,E)$, the cut polytope $\mathrm{CUT}(G)$ is defined as the convex hull of the incidence vectors of all cuts in $G$. The 1-skeleton of $\mathrm{CUT}(G)$ is a graph whose vertex set is the vertex set of the polytope, and the edge set is the set of geometric edges or one-dimensional faces of the polytope. We study the diameter and the clique number of 1-skeleton of cu…
▽ More
Given an undirected graph $G = (V,E)$, the cut polytope $\mathrm{CUT}(G)$ is defined as the convex hull of the incidence vectors of all cuts in $G$. The 1-skeleton of $\mathrm{CUT}(G)$ is a graph whose vertex set is the vertex set of the polytope, and the edge set is the set of geometric edges or one-dimensional faces of the polytope. We study the diameter and the clique number of 1-skeleton of cut polytopes for several classes of graphs. These characteristics are of interest since they estimate the computational complexity of the max-cut problem for certain computational models and classes of algorithms. It is established that while the diameter of the 1-skeleton of a cut polytope does not exceed $|V|-1$ for any connected graph, the clique number varies significantly depending on the class of graphs. For trees, cacti, and almost trees (2), the clique number is linear in the dimension, whereas for complete bipartite and $k$-partite graphs, it is superpolynomial.
△ Less
Submitted 21 February, 2024;
originally announced February 2024.
-
Artin $L$-functions and noncommutative tori
Authors:
Igor V. Nikolaev
Abstract:
Using the ideas of Deninger, we prove that the Artin $L$-functions coincide with such of the noncommutative tori. This result can be viewed as the Langlands reciprocity for noncommutative tori.
Using the ideas of Deninger, we prove that the Artin $L$-functions coincide with such of the noncommutative tori. This result can be viewed as the Langlands reciprocity for noncommutative tori.
△ Less
Submitted 29 December, 2023;
originally announced December 2023.
-
Energy barriers of Be and B in passing through the C60 fullerene cage
Authors:
A. V. Bibikov,
A. V. Nikolaev,
P. V. Borisyuk,
E. V. Tkalya
Abstract:
We have studied the potential barriers for the penetration of atomic beryllium or boron inside the C60 fullerene by performing ab initio density functional theory (DFT) calculations with three variants for the exchange and correlation: B3LYP (hybrid functional), PW91 and PBE. Four principal trajectories to the inner part of C60 for the penetrating atom have been considered: through the center of s…
▽ More
We have studied the potential barriers for the penetration of atomic beryllium or boron inside the C60 fullerene by performing ab initio density functional theory (DFT) calculations with three variants for the exchange and correlation: B3LYP (hybrid functional), PW91 and PBE. Four principal trajectories to the inner part of C60 for the penetrating atom have been considered: through the center of six-member-carbon ring (hexagon), five-member-carbon ring (pentagon), and also through the center of the double C-C bond (D-bond) and the center of the single C-C bond (S-bond). Averaging over the three DFT variants yields the following barriers for beryllium penetrating inside a deformable fullerene: 3.2 eV (hexagon), 4.8 eV (S-bond), 5.3 eV (D-bond), 5.9~eV (pentagon). These barriers correspond to the slow and adiabatic penetration of Be, in contrast to the fast (non-adiabatic) penetration through the rigid cage of C60 resulting in 5.6 eV (hexagon), 16.3 eV (pentagon), 81.8 eV (S-bond) and 93.4 eV (D-bond). The potential barriers for the boron penetrating inside deformable/rigid C60 are: 3.7/105.4 eV (D-bond), 4.0/86.8 eV (S-bond), 4.7/7.8 eV (hexagon), 6.8/14.0 eV (pentagon). The potential barriers for Be and B esca** from the inner part of C$_{60}$ are higher by the value of 0.84 eV for Be and 0.81 eV for B. The considerable reduction of the potential barriers for the deformable fullerene is ascribed to the formation of the Be-C and B-C bonds. We discuss the difference between Be and B, compare three variants of DFT, and analyze the role of the dispersion interaction.
△ Less
Submitted 21 December, 2023;
originally announced December 2023.
-
Gemini: A Family of Highly Capable Multimodal Models
Authors:
Gemini Team,
Rohan Anil,
Sebastian Borgeaud,
Jean-Baptiste Alayrac,
Jiahui Yu,
Radu Soricut,
Johan Schalkwyk,
Andrew M. Dai,
Anja Hauth,
Katie Millican,
David Silver,
Melvin Johnson,
Ioannis Antonoglou,
Julian Schrittwieser,
Amelia Glaese,
Jilin Chen,
Emily Pitler,
Timothy Lillicrap,
Angeliki Lazaridou,
Orhan Firat,
James Molloy,
Michael Isard,
Paul R. Barham,
Tom Hennigan,
Benjamin Lee
, et al. (1325 additional authors not shown)
Abstract:
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultr…
▽ More
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.
△ Less
Submitted 17 June, 2024; v1 submitted 18 December, 2023;
originally announced December 2023.
-
From Transcripts to Insights: Uncovering Corporate Risks Using Generative AI
Authors:
Alex Kim,
Maximilian Muhn,
Valeri Nikolaev
Abstract:
We explore the value of generative AI tools, such as ChatGPT, in hel** investors uncover dimensions of corporate risk. We develop and validate firm-level measures of risk exposure to political, climate, and AI-related risks. Using the GPT 3.5 model to generate risk summaries and assessments from the context provided by earnings call transcripts, we show that GPT-based measures possess significan…
▽ More
We explore the value of generative AI tools, such as ChatGPT, in hel** investors uncover dimensions of corporate risk. We develop and validate firm-level measures of risk exposure to political, climate, and AI-related risks. Using the GPT 3.5 model to generate risk summaries and assessments from the context provided by earnings call transcripts, we show that GPT-based measures possess significant information content and outperform the existing risk measures in predicting (abnormal) firm-level volatility and firms' choices such as investment and innovation. Importantly, information in risk assessments dominates that in risk summaries, establishing the value of general AI knowledge. We also find that generative AI is effective at detecting emerging risks, such as AI risk, which has soared in recent quarters. Our measures perform well both within and outside the GPT's training window and are priced in equity markets. Taken together, an AI-based approach to risk measurement provides useful insights to users of corporate disclosures at a low cost.
△ Less
Submitted 26 October, 2023;
originally announced October 2023.
-
Non-abelian class field theory and higher dimensional noncommutative tori
Authors:
Igor V. Nikolaev
Abstract:
We study a relation between the Drinfeld modules and the even dimensional noncommutative tori. A non-abelian class field theory is developed based on this relation. Explicit generators of the Galois extensions are constructed.
We study a relation between the Drinfeld modules and the even dimensional noncommutative tori. A non-abelian class field theory is developed based on this relation. Explicit generators of the Galois extensions are constructed.
△ Less
Submitted 11 September, 2023;
originally announced September 2023.
-
Bloated Disclosures: Can ChatGPT Help Investors Process Information?
Authors:
Alex Kim,
Maximilian Muhn,
Valeri Nikolaev
Abstract:
Generative AI tools such as ChatGPT can fundamentally change the way investors process information. We probe the economic usefulness of these tools in summarizing complex corporate disclosures using the stock market as a laboratory. The unconstrained summaries are remarkably shorter compared to the originals, whereas their information content is amplified. When a document has a positive (negative)…
▽ More
Generative AI tools such as ChatGPT can fundamentally change the way investors process information. We probe the economic usefulness of these tools in summarizing complex corporate disclosures using the stock market as a laboratory. The unconstrained summaries are remarkably shorter compared to the originals, whereas their information content is amplified. When a document has a positive (negative) sentiment, its summary becomes more positive (negative). Importantly, the summaries are more effective at explaining stock market reactions to the disclosed information. Motivated by these findings, we propose a measure of information ``bloat." We show that bloated disclosure is associated with adverse capital market consequences, such as lower price efficiency and higher information asymmetry. Finally, we show that the model is effective at constructing targeted summaries that identify firms' (non-)financial performance. Collectively, our results indicate that generative AI adds considerable value for investors with information processing constraints.
△ Less
Submitted 3 February, 2024; v1 submitted 16 June, 2023;
originally announced June 2023.
-
Drinfeld modules as noncommutative tori
Authors:
Igor V. Nikolaev
Abstract:
The Drinfeld module is a tool of the explicit class field theory for the function fields. We first observe a similarity of such modules with the noncommutative tori, and then use it to develop an explicit class field theory for the number fields. The case of the imaginary quadratic number fields is treated in detail.
The Drinfeld module is a tool of the explicit class field theory for the function fields. We first observe a similarity of such modules with the noncommutative tori, and then use it to develop an explicit class field theory for the number fields. The case of the imaginary quadratic number fields is treated in detail.
△ Less
Submitted 28 January, 2024; v1 submitted 10 June, 2023;
originally announced June 2023.
-
SEAHORSE: A Multilingual, Multifaceted Dataset for Summarization Evaluation
Authors:
Elizabeth Clark,
Shruti Rijhwani,
Sebastian Gehrmann,
Joshua Maynez,
Roee Aharoni,
Vitaly Nikolaev,
Thibault Sellam,
Aditya Siddhant,
Dipanjan Das,
Ankur P. Parikh
Abstract:
Reliable automatic evaluation of summarization systems is challenging due to the multifaceted and subjective nature of the task. This is especially the case for languages other than English, where human evaluations are scarce. In this work, we introduce SEAHORSE, a dataset for multilingual, multifaceted summarization evaluation. SEAHORSE consists of 96K summaries with human ratings along 6 dimensi…
▽ More
Reliable automatic evaluation of summarization systems is challenging due to the multifaceted and subjective nature of the task. This is especially the case for languages other than English, where human evaluations are scarce. In this work, we introduce SEAHORSE, a dataset for multilingual, multifaceted summarization evaluation. SEAHORSE consists of 96K summaries with human ratings along 6 dimensions of text quality: comprehensibility, repetition, grammar, attribution, main ideas, and conciseness, covering 6 languages, 9 systems and 4 datasets. As a result of its size and scope, SEAHORSE can serve both as a benchmark to evaluate learnt metrics, as well as a large-scale resource for training such metrics. We show that metrics trained with SEAHORSE achieve strong performance on the out-of-domain meta-evaluation benchmarks TRUE (Honovich et al., 2022) and mFACE (Aharoni et al., 2022). We make the SEAHORSE dataset and metrics publicly available for future research on multilingual and multifaceted summarization evaluation.
△ Less
Submitted 1 November, 2023; v1 submitted 22 May, 2023;
originally announced May 2023.
-
One- and two-particle correlation functions in the cluster perturbation theory for cuprates
Authors:
V. I. Kuz'min,
S. V. Nikolaev,
M. M. Korshunov,
S. G. Ovchinnikov
Abstract:
Physics of high-$T_c$ superconducting cuprates is obscured by the effect of strong electronic correlations. One way to overcome the problem is to seek for an exact solution at least within the small cluster and expand it to the whole crystal. Such an approach is in the heart of the cluster perturbation theory (CPT). Here we develop CPT for the dynamic spin and charge susceptibilities (spin-CPT and…
▽ More
Physics of high-$T_c$ superconducting cuprates is obscured by the effect of strong electronic correlations. One way to overcome the problem is to seek for an exact solution at least within the small cluster and expand it to the whole crystal. Such an approach is in the heart of the cluster perturbation theory (CPT). Here we develop CPT for the dynamic spin and charge susceptibilities (spin-CPT and charge-CPT), within which the correlation effects are explicitly taken into account by the exact diagonalization. We apply spin-CPT and charge-CPT to the effective two-band Hubbard model for the cuprates obtained from the three-band Emery model and calculate one- and two-particle correlation functions, namely, spectral function and spin and charge susceptibilities. Do** dependence of the spin susceptibility was studied within spin-CPT and CPT-RPA that is the CPT generalization of the random phase approximation (RPA). Both methods produce the low energy response at four incommensurate wave vectors in qualitative agreement to the results of the inelastic neutron scattering on overdoped cuprates.
△ Less
Submitted 3 May, 2023;
originally announced May 2023.
-
Remark on function field analogy
Authors:
Igor V. Nikolaev
Abstract:
We study the analogy between number fields and function fields in one variable over finite fields. The main result is an isomorphism between the Hilbert class fields of class number one and a family of the function fields $\mathbf{F}_q(C)$ over a desingularized algebraic curve $C$. Our proof is based on the K-theory of the Serre $C^*$-algebras and birational geometry of the curve $C$. We apply the…
▽ More
We study the analogy between number fields and function fields in one variable over finite fields. The main result is an isomorphism between the Hilbert class fields of class number one and a family of the function fields $\mathbf{F}_q(C)$ over a desingularized algebraic curve $C$. Our proof is based on the K-theory of the Serre $C^*$-algebras and birational geometry of the curve $C$. We apply the isomorphism to construct explicit generators of the Hilbert class fields coming from the torsion submodules of the Drinfeld module.
△ Less
Submitted 24 February, 2023;
originally announced February 2023.
-
On cone partitions for the min-cut and max-cut problems with non-negative edges
Authors:
Andrei V. Nikolaev,
Alexander V. Korostil
Abstract:
We consider the classical minimum and maximum cut problems: find a partition of vertices of a graph into two disjoint subsets that minimize or maximize the sum of the weights of edges with endpoints in different subsets. It is known that if the edge weights are non-negative, then the min-cut problem is polynomially solvable, while the max-cut problem is NP-hard.
We construct a partition of the p…
▽ More
We consider the classical minimum and maximum cut problems: find a partition of vertices of a graph into two disjoint subsets that minimize or maximize the sum of the weights of edges with endpoints in different subsets. It is known that if the edge weights are non-negative, then the min-cut problem is polynomially solvable, while the max-cut problem is NP-hard.
We construct a partition of the positive orthant into convex cones corresponding to the characteristic cut vectors, similar to a normal fan of a cut polyhedron. A graph of a cone partition is a graph whose vertices are cones, and two cones are adjacent if and only if they have a common facet. We define adjacency criteria in the graphs of cone partitions for the min-cut and max-cut problems. Based on them, we show that for both problems the vertex degrees are exponential, and the graph diameter equals 2. These results contrast with the clique numbers of graphs of cone partitions, which are linear for the minimum cut problem and exponential for the maximum cut problem.
△ Less
Submitted 11 May, 2023; v1 submitted 8 February, 2023;
originally announced February 2023.
-
Birational geometry of quaternions
Authors:
Igor Nikolaev
Abstract:
The Hilbert class field of the quaternion algebra $B$ is an algebra $\mathscr{H}(B)$ such that every two-sided ideal of $B$ is principal in $\mathscr{H}(B)$. We study the avatars of $B$ and $\mathscr{H}(B)$, i.e. algebraic surfaces attached to the quaternion algebras. It is proved that the avatar of $\mathscr{H}(B)$ is obtained from the avatar of $B$ by a birational map (blow-up). We apply this re…
▽ More
The Hilbert class field of the quaternion algebra $B$ is an algebra $\mathscr{H}(B)$ such that every two-sided ideal of $B$ is principal in $\mathscr{H}(B)$. We study the avatars of $B$ and $\mathscr{H}(B)$, i.e. algebraic surfaces attached to the quaternion algebras. It is proved that the avatar of $\mathscr{H}(B)$ is obtained from the avatar of $B$ by a birational map (blow-up). We apply this result to the function field analogy.
△ Less
Submitted 13 December, 2022;
originally announced December 2022.
-
Photoinduced Magnetic Transitions and Excitonic Order Enhancement in Spin Crossover Strongly Correlated Electron Systems
Authors:
Yuri. S. Orlov,
Sergey. V. Nikolaev
Abstract:
The effects associated with exciton Bose condensate formation in strongly correlated spin crossover systems are considered within the effective Hamiltonian obtained from the two-orbital Hubbard-Kanamori model. The collective excitations spectrum at various points of the temperature-crystal field phase diagram is calculated. The role of the electron-phonon interaction is discussed. The exciton and…
▽ More
The effects associated with exciton Bose condensate formation in strongly correlated spin crossover systems are considered within the effective Hamiltonian obtained from the two-orbital Hubbard-Kanamori model. The collective excitations spectrum at various points of the temperature-crystal field phase diagram is calculated. The role of the electron-phonon interaction is discussed. The exciton and magnetic order photoenhancement (induction) in strongly correlated spin crossover systems a new mechanism based on the cooperative effect of electron-phonon and interatomic exchange interactions and the appearance of a massive collective phase mode is demonstrated.
△ Less
Submitted 12 December, 2022;
originally announced December 2022.
-
Reply to Comment on "Multiple locations of boron atoms in the exohedral and endohedral C60 fullerene" by J. Xu and G.-L. Hou
Authors:
A. V. Bibikov,
A. V. Nikolaev,
I. V. Bodrenko,
P. V. Borisyuk,
E. V. Tkalya
Abstract:
In three out of five cases considered in our work, DFT calculations presented by Xu and Hou in their Comment give the same ground state confirmations. On the other hand, depending on the choice of the exchange-correlation functional, the geometry optimization within DFT results in different ground state confirmations for B@C60 and B60, Table I of the Comment. Therefore, the energy balance between…
▽ More
In three out of five cases considered in our work, DFT calculations presented by Xu and Hou in their Comment give the same ground state confirmations. On the other hand, depending on the choice of the exchange-correlation functional, the geometry optimization within DFT results in different ground state confirmations for B@C60 and B60, Table I of the Comment. Therefore, the energy balance between nearest confirmations in these molecular complexes is subtle, and various methods can give different ground state structures. Consequently, the results of our method - the Hartree-Fock (HF) approach with the second order Møller-Plesset perturbation theory (MP2) - should be compared with the DFT results on equal ground, we cannot agree that the DFT method used in the Comment is superior to HF-MP2. In the Reply, we also present additional HF calculations with the 6-31G* basis set (used in the Comment for the geometry optimization) to show that the polarization functions do not change the ground state confirmations obtained by us earlier at the HF/6-31G level.
△ Less
Submitted 11 November, 2022;
originally announced November 2022.
-
Light-induced ultrafast dynamics of spin crossovers in LaCoO3
Authors:
Yu. S. Orlov,
S. V. Nikolaev,
S. G. Ovchinnikov
Abstract:
Ultrafast quantum dynamics relaxation of a photoexcited state in a strongly correlated spin crossover system LaCoO3 under a sudden perturbation is considered with the density matrix generalized master equation. The magnetization and cobalt-oxygen bond length oscillations were found. The evolution of the electronic band structure during relaxation is calculated in the framework of the LDA+GTB metho…
▽ More
Ultrafast quantum dynamics relaxation of a photoexcited state in a strongly correlated spin crossover system LaCoO3 under a sudden perturbation is considered with the density matrix generalized master equation. The magnetization and cobalt-oxygen bond length oscillations were found. The evolution of the electronic band structure during relaxation is calculated in the framework of the LDA+GTB method.
△ Less
Submitted 2 November, 2022;
originally announced November 2022.
-
TaTa: A Multilingual Table-to-Text Dataset for African Languages
Authors:
Sebastian Gehrmann,
Sebastian Ruder,
Vitaly Nikolaev,
Jan A. Botha,
Michael Chavinda,
Ankur Parikh,
Clara Rivera
Abstract:
Existing data-to-text generation datasets are mostly limited to English. To address this lack of data, we create Table-to-Text in African languages (TaTa), the first large multilingual table-to-text dataset with a focus on African languages. We created TaTa by transcribing figures and accompanying text in bilingual reports by the Demographic and Health Surveys Program, followed by professional tra…
▽ More
Existing data-to-text generation datasets are mostly limited to English. To address this lack of data, we create Table-to-Text in African languages (TaTa), the first large multilingual table-to-text dataset with a focus on African languages. We created TaTa by transcribing figures and accompanying text in bilingual reports by the Demographic and Health Surveys Program, followed by professional translation to make the dataset fully parallel. TaTa includes 8,700 examples in nine languages including four African languages (Hausa, Igbo, Swahili, and Yorùbá) and a zero-shot test language (Russian). We additionally release screenshots of the original figures for future research on multilingual multi-modal approaches. Through an in-depth human evaluation, we show that TaTa is challenging for current models and that less than half the outputs from an mT5-XXL-based model are understandable and attributable to the source data. We further demonstrate that existing metrics perform poorly for TaTa and introduce learned metrics that achieve a high correlation with human judgments. We release all data and annotations at https://github.com/google-research/url-nlp.
△ Less
Submitted 31 October, 2022;
originally announced November 2022.
-
Geometry of division rings
Authors:
Igor Nikolaev
Abstract:
We prove an analog of Belyi's theorem for the algebraic surfaces. Namely, any non-singular algebraic surface can be defined over a number field if and only it covers the complex projective plane with ramification at three knotted two-dimensional spheres.
We prove an analog of Belyi's theorem for the algebraic surfaces. Namely, any non-singular algebraic surface can be defined over a number field if and only it covers the complex projective plane with ramification at three knotted two-dimensional spheres.
△ Less
Submitted 13 September, 2022;
originally announced September 2022.
-
Self-Sustained Non-Equilibrium Co-existence of Fluid and Solid States in a Strongly Coupled Complex Plasma System
Authors:
M. G Hariprasad,
P. Bandyopadhyay,
V. S. Nikolaev,
D. A. Kolotinskii,
S. Arumugam,
G. Arora,
S. Singh,
A. Sen,
A. V. Timofeev
Abstract:
A complex (dusty) plasma system is well known as a paradigmatic model for studying the kinetics of solid-liquid phase transitions in inactive condensed matter. At the same time, under certain conditions a complex plasma system can also display characteristics of an active medium with the micron-sized particles converting energy of the ambient environment into motility and thereby becoming active.…
▽ More
A complex (dusty) plasma system is well known as a paradigmatic model for studying the kinetics of solid-liquid phase transitions in inactive condensed matter. At the same time, under certain conditions a complex plasma system can also display characteristics of an active medium with the micron-sized particles converting energy of the ambient environment into motility and thereby becoming active. We present a detailed analysis of the experimental complex plasmas system that shows evidence of a non-equilibrium stationary coexistence between a cold crystalline and a hot fluid state in the structure due to the conversion of plasma energy into the motion energy of microparticles in the central region of the system. The plasma mediated non-reciprocal interaction between the dust particles is the underlying mechanism for the enormous heating of the central subsystem, and it acts as a micro-scale energy source that keeps the central subsystem in the molten state. Accurate multiscale simulations of the system based on combined molecular dynamics and particle-in-cell approaches show that strong structural nonuniformity of the system under the action of electostatic trap makes development of instabilities a local process. We present both experimental tests conducted with a complex plasmas system in a DC glow discharge plasma and a detailed theoretical analysis.
△ Less
Submitted 13 August, 2022;
originally announced August 2022.
-
Excitonic ordering in strongly correlated spin crossover systems: induced magnetism and excitonic excitation spectrum
Authors:
Yu. S. Orlov,
S. V. Nikolaev,
V. I. Kuz'min,
A. E. Zarubin,
S. G. Ovchinnikov
Abstract:
The effects associated with interatomic hop**s of excitons and the excitonic Bose condensate formation in the strongly correlated spin crossover systems are considered in the framework of the effective Hamiltonian for the two-band Kanamori model. The appearance of antiferromagnetic ordering due to the exciton order is found even in the absence of interatomic exchange interaction. The spectrum of…
▽ More
The effects associated with interatomic hop**s of excitons and the excitonic Bose condensate formation in the strongly correlated spin crossover systems are considered in the framework of the effective Hamiltonian for the two-band Kanamori model. The appearance of antiferromagnetic ordering due to the exciton order is found even in the absence of interatomic exchange interaction. The spectrum of excitonic excitations is calculated at various points of the "temperature vs. crystal field" phase diagram. Outside the region of exciton ordering, the spectrum has a gap, which vanishes at the boundary of the exciton condensate phase. The non-uniform spectral weight distribution over the Brillouin zone is found. The role of electron-phonon interaction is discussed as well.
△ Less
Submitted 28 June, 2022;
originally announced June 2022.
-
GEMv2: Multilingual NLG Benchmarking in a Single Line of Code
Authors:
Sebastian Gehrmann,
Abhik Bhattacharjee,
Abinaya Mahendiran,
Alex Wang,
Alexandros Papangelis,
Aman Madaan,
Angelina McMillan-Major,
Anna Shvets,
Ashish Upadhyay,
Bingsheng Yao,
Bryan Wilie,
Chandra Bhagavatula,
Chaobin You,
Craig Thomson,
Cristina Garbacea,
Dakuo Wang,
Daniel Deutsch,
Deyi Xiong,
Di **,
Dimitra Gkatzia,
Dragomir Radev,
Elizabeth Clark,
Esin Durmus,
Faisal Ladhak,
Filip Ginter
, et al. (52 additional authors not shown)
Abstract:
Evaluation in machine learning is usually informed by past choices, for example which datasets or metrics to use. This standardization enables the comparison on equal footing using leaderboards, but the evaluation choices become sub-optimal as better alternatives arise. This problem is especially pertinent in natural language generation which requires ever-improving suites of datasets, metrics, an…
▽ More
Evaluation in machine learning is usually informed by past choices, for example which datasets or metrics to use. This standardization enables the comparison on equal footing using leaderboards, but the evaluation choices become sub-optimal as better alternatives arise. This problem is especially pertinent in natural language generation which requires ever-improving suites of datasets, metrics, and human evaluation to make definitive claims. To make following best model evaluation practices easier, we introduce GEMv2. The new version of the Generation, Evaluation, and Metrics Benchmark introduces a modular infrastructure for dataset, model, and metric developers to benefit from each others work. GEMv2 supports 40 documented datasets in 51 languages. Models for all datasets can be evaluated online and our interactive data card creation and rendering tools make it easier to add new datasets to the living benchmark.
△ Less
Submitted 24 June, 2022; v1 submitted 22 June, 2022;
originally announced June 2022.
-
Remark on ordered braid groups
Authors:
Igor Nikolaev
Abstract:
We recover the Dehornoy order on the braid group $B_{2g+n}$ from the tracial state on a cluster $C^*$-algebra $\mathbb{A}(S_{g,n})$ associated to the surface $S_{g,n}$ of genus $g$ with $n$ boundary components. It is proved that the space of left-ordering of the fundamental group $π_1(S_{g,n})$ is a totally disconnected dense subspace of the projective Teichmüller space…
▽ More
We recover the Dehornoy order on the braid group $B_{2g+n}$ from the tracial state on a cluster $C^*$-algebra $\mathbb{A}(S_{g,n})$ associated to the surface $S_{g,n}$ of genus $g$ with $n$ boundary components. It is proved that the space of left-ordering of the fundamental group $π_1(S_{g,n})$ is a totally disconnected dense subspace of the projective Teichmüller space $\mathbb{P}T_{g,n}\cong \mathbf{R}^{6g-7+2n}$. In particular, each left-ordering of $π_1(S_{g,n})$ defines the orbit of a Riemann surface $S_{g,n}$ under the geodesic flow on the space $T_{g,n}$.
△ Less
Submitted 11 May, 2023; v1 submitted 16 June, 2022;
originally announced June 2022.
-
Aspects of Superdeterminism Made Intuitive
Authors:
Vitaly Nikolaev,
Louis Vervoort
Abstract:
We attempt to make superdeterminism more intuitive, notably by simulating a deterministic model system, a billiard game. In this system an initial 'bang' correlates all events, just as in the superdeterministic universe. We introduce the notions of 'strong' and 'soft' superdeterminism, in order to clarify debates in the literature. Based on the analogy with billiards, we show that superdeterminist…
▽ More
We attempt to make superdeterminism more intuitive, notably by simulating a deterministic model system, a billiard game. In this system an initial 'bang' correlates all events, just as in the superdeterministic universe. We introduce the notions of 'strong' and 'soft' superdeterminism, in order to clarify debates in the literature. Based on the analogy with billiards, we show that superdeterministic correlations may exist as a matter of principle, but be undetectable for all practical purposes. This allows us to counter classical objections to superdeterminism such as the claim that it would be at odds with the scientific method, and with the construction of new theories. Finally, we show that probability theory, as a physical theory, indicates that superdeterminism has a greater explanatory power than its competitors: it can coherently answer questions for which other positions remain powerless.
△ Less
Submitted 21 May, 2022;
originally announced May 2022.
-
Possible quadrupole-order-driven commensurate-incommensurate phase transition in B20 CoGe
Authors:
S. -H. Baek,
V. A. Sidorov,
A. V. Nikolaev,
T. Klimczuk,
F. Ronning,
A. V. Tsvyashchenko
Abstract:
The B20-type cobalt germanide CoGe was investigated by measuring the specific heat, resistivity, and $^{59}$Co nuclear magnetic resonance (NMR). We observed a phase transition at $T_Q=13.7$ K, evidenced by a very narrow peak of the specific heat and sharp changes of the nuclear spin-spin ($T_2^{-1}$) and spin-lattice ($T_1^{-1}$) relaxation rates. The fact that the entropy release is extremely sma…
▽ More
The B20-type cobalt germanide CoGe was investigated by measuring the specific heat, resistivity, and $^{59}$Co nuclear magnetic resonance (NMR). We observed a phase transition at $T_Q=13.7$ K, evidenced by a very narrow peak of the specific heat and sharp changes of the nuclear spin-spin ($T_2^{-1}$) and spin-lattice ($T_1^{-1}$) relaxation rates. The fact that the entropy release is extremely small and the Knight shift is almost independent of temperature down to low temperatures as anticipated in a paramagnetic metal indicates that the $T_Q$ transition is of non-magnetic origin. In addition, we detected a crossover scale $T_0\sim30$ K below which the resistivity and the NMR linewidth increase, and $T_1^{-1}$ is progressively distributed in space, that is, a static and dynamical spatial inhomogeneity develops. While the order parameter for the $T_Q$ transition remains an open question, a group-theoretical analysis suggests that the finite electric quadrupole density arising from the low local site symmetry at cobalt sites could drive the crystal symmetry lowering from the P2$_1$3 symmetry that is commensurate to the R3 symmetry with an incommensurate wavevector, which fairly well accounts for the $T_Q$ transition. The quadrupole-order-driven commensurate-incommensurate phase transition may be another remarkable phenomenon arising from the structural chirality inherent in the noncentrosymmetric B20 family.
△ Less
Submitted 18 April, 2022;
originally announced April 2022.
-
Comments on a note by M. Waldschmidt
Authors:
Igor Nikolaev
Abstract:
This note is the follow up to a paper by M. Waldschmidt.
This note is the follow up to a paper by M. Waldschmidt.
△ Less
Submitted 26 March, 2022;
originally announced March 2022.
-
Finding a second Hamiltonian decomposition of a 4-regular multigraph by integer linear programming
Authors:
Andrei V. Nikolaev,
Egor V. Klimov
Abstract:
A Hamiltonian decomposition of a regular graph is a partition of its edge set into Hamiltonian cycles. We consider the second Hamiltonian decomposition problem: for a 4-regular multigraph find 2 edge-disjoint Hamiltonian cycles different from the given ones. This problem arises in polyhedral combinatorics as a sufficient condition for non-adjacency in the 1-skeleton of the travelling salesperson p…
▽ More
A Hamiltonian decomposition of a regular graph is a partition of its edge set into Hamiltonian cycles. We consider the second Hamiltonian decomposition problem: for a 4-regular multigraph find 2 edge-disjoint Hamiltonian cycles different from the given ones. This problem arises in polyhedral combinatorics as a sufficient condition for non-adjacency in the 1-skeleton of the travelling salesperson polytope. We introduce two integer linear programming models for the problem based on the classical Dantzig-Fulkerson-Johnson and Miller-Tucker-Zemlin formulations for the travelling salesperson problem. To enhance the performance on feasible problems, we supplement the algorithm with a variable neighbourhood descent heuristic w.r.t. two neighbourhood structures, and a chain edge fixing procedure. Based on the computational experiments, the Dantzig-Fulkerson-Johnson formulation showed the best results on directed multigraphs, while on undirected multigraphs, the variable neighbourhood descent heuristic was especially effective.
△ Less
Submitted 11 January, 2022;
originally announced January 2022.
-
Measuring Attribution in Natural Language Generation Models
Authors:
Hannah Rashkin,
Vitaly Nikolaev,
Matthew Lamm,
Lora Aroyo,
Michael Collins,
Dipanjan Das,
Slav Petrov,
Gaurav Singh Tomar,
Iulia Turc,
David Reitter
Abstract:
With recent improvements in natural language generation (NLG) models for various applications, it has become imperative to have the means to identify and evaluate whether NLG output is only sharing verifiable information about the external world. In this work, we present a new evaluation framework entitled Attributable to Identified Sources (AIS) for assessing the output of natural language genera…
▽ More
With recent improvements in natural language generation (NLG) models for various applications, it has become imperative to have the means to identify and evaluate whether NLG output is only sharing verifiable information about the external world. In this work, we present a new evaluation framework entitled Attributable to Identified Sources (AIS) for assessing the output of natural language generation models, when such output pertains to the external world. We first define AIS and introduce a two-stage annotation pipeline for allowing annotators to appropriately evaluate model output according to AIS guidelines. We empirically validate this approach on generation datasets spanning three tasks (two conversational QA datasets, a summarization dataset, and a table-to-text dataset) via human evaluation studies that suggest that AIS could serve as a common framework for measuring whether model-generated statements are supported by underlying sources. We release guidelines for the human evaluation studies.
△ Less
Submitted 2 August, 2022; v1 submitted 23 December, 2021;
originally announced December 2021.
-
Quantum arithmetic dynamics
Authors:
Igor Nikolaev
Abstract:
We study dynamics of the Lattès maps in the complex plane in terms of the Cuntz-Krieger algebras associated to the endomorphisms of the non-commutative tori. In particular, it is shown that iterations of the Lattès maps can be reduced to the dynamics of the subshifts of finite type. Using such a reduction, we calculate the zeta function of the Lattès maps.
We study dynamics of the Lattès maps in the complex plane in terms of the Cuntz-Krieger algebras associated to the endomorphisms of the non-commutative tori. In particular, it is shown that iterations of the Lattès maps can be reduced to the dynamics of the subshifts of finite type. Using such a reduction, we calculate the zeta function of the Lattès maps.
△ Less
Submitted 21 December, 2021;
originally announced December 2021.
-
SynthBio: A Case Study in Human-AI Collaborative Curation of Text Datasets
Authors:
Ann Yuan,
Daphne Ippolito,
Vitaly Nikolaev,
Chris Callison-Burch,
Andy Coenen,
Sebastian Gehrmann
Abstract:
NLP researchers need more, higher-quality text datasets. Human-labeled datasets are expensive to collect, while datasets collected via automatic retrieval from the web such as WikiBio are noisy and can include undesired biases. Moreover, data sourced from the web is often included in datasets used to pretrain models, leading to inadvertent cross-contamination of training and test sets. In this wor…
▽ More
NLP researchers need more, higher-quality text datasets. Human-labeled datasets are expensive to collect, while datasets collected via automatic retrieval from the web such as WikiBio are noisy and can include undesired biases. Moreover, data sourced from the web is often included in datasets used to pretrain models, leading to inadvertent cross-contamination of training and test sets. In this work we introduce a novel method for efficient dataset curation: we use a large language model to provide seed generations to human raters, thereby changing dataset authoring from a writing task to an editing task. We use our method to curate SynthBio - a new evaluation set for WikiBio - composed of structured attribute lists describing fictional individuals, mapped to natural language biographies. We show that our dataset of fictional biographies is less noisy than WikiBio, and also more balanced with respect to gender and nationality.
△ Less
Submitted 12 January, 2022; v1 submitted 11 November, 2021;
originally announced November 2021.
-
K-theory of Jones polynomials
Authors:
Andrey Glubokov,
Igor Nikolaev
Abstract:
We recover the Jones polynomials of knots and links from the K-theory of a cluster C*-algebra of the sphere with two cusps. In particular, an interplay between the Chebyshev and Jones polynomials is studied.
We recover the Jones polynomials of knots and links from the K-theory of a cluster C*-algebra of the sphere with two cusps. In particular, an interplay between the Chebyshev and Jones polynomials is studied.
△ Less
Submitted 19 September, 2022; v1 submitted 11 October, 2021;
originally announced October 2021.
-
Class field towers and minimal models
Authors:
Igor Nikolaev
Abstract:
It is shown that the real class field towers are always finite. The proof is based on Castelnuovo's theory of the algebraic surfaces and a functor from such surfaces to the Etesi C*-algebras.
It is shown that the real class field towers are always finite. The proof is based on Castelnuovo's theory of the algebraic surfaces and a functor from such surfaces to the Etesi C*-algebras.
△ Less
Submitted 30 August, 2021;
originally announced August 2021.
-
Highly Boron-Doped Graphite and Diamond Synthesized From Adamantane and Ortho-Carborane under High Pressure
Authors:
Rustem H. Bagramov,
Vladimir P. Filonenko,
Igor P. Zibrov,
Elena A. Skryleva,
Alexander V. Nikolaev,
Dmitrii G. Pasternak,
Igor I. Vlasov
Abstract:
This work demonstrates the effectiveness of the high-pressure method for the production of graphite and diamond with a high degree of boron do** using adamantanecarborane mixture as a precursor. At 8 GPa and $1700 ^{o}C$, graphite is obtained from adamantane $C_{10}H_{16}$, whereas microcrystals of boron-doped diamond (2÷2.5 at.% of boron) are synthesized from a mixture of adamantane and ortho-c…
▽ More
This work demonstrates the effectiveness of the high-pressure method for the production of graphite and diamond with a high degree of boron do** using adamantanecarborane mixture as a precursor. At 8 GPa and $1700 ^{o}C$, graphite is obtained from adamantane $C_{10}H_{16}$, whereas microcrystals of boron-doped diamond (2÷2.5 at.% of boron) are synthesized from a mixture of adamantane and ortho-carborane $C_{2}B_{10}H_{12}$ (atomic ratio B:C = 5:95). This result shows convincingly the catalytical activity of boron in the synthesis of diamond under high pressure. At pressures lower than 7 GPa, only graphite is synthesized from the adamantane and carborane mixture. Graphitization starts at quite low temperatures (below $1400 ^{o}C$) and an increase in temperature simultaneously increases boron content and the quality of the graphite crystal lattice. Thorough study of the material structure allows us to assume that the substitutional boron atoms are distributed periodically and equidistantly from each other in the graphite layers at high boron concentrations (>1 at.%). The theoretical arguments and model ab initio calculations confirm this assumption and explain the experimentally observed boron concentrations.
△ Less
Submitted 10 August, 2021;
originally announced August 2021.
-
Noncommutative geometry of elliptic surfaces
Authors:
Igor Nikolaev
Abstract:
We recast elliptic surfaces over the projective line in terms of the non-commutative tori and one-parameter families of the periodic continued fractions. The correspondence is used to study the Picard numbers, the ranks and the minimal models of such surfaces. As an example, we calculate the Picard numbers of elliptic surfaces with fibers having complex multiplication.
We recast elliptic surfaces over the projective line in terms of the non-commutative tori and one-parameter families of the periodic continued fractions. The correspondence is used to study the Picard numbers, the ranks and the minimal models of such surfaces. As an example, we calculate the Picard numbers of elliptic surfaces with fibers having complex multiplication.
△ Less
Submitted 26 April, 2024; v1 submitted 21 June, 2021;
originally announced June 2021.
-
HVPE growth of corundum-structured $α$-Ga$_2$O$_3$ on sapphire substrates with $α$-Cr$_2$O$_3$ buffer layer
Authors:
S. I. Stepanov,
V. I. Nikolaev,
A. V. Almaev,
A. I. Pechnikov,
M. P. Scheglov,
A. V. Chikiryaka,
B. O. Kushnarev
Abstract:
Gallium oxide films were grown by HVPE on (0001) sapphire substrates with and without $α$-Cr$_2$O$_3$ buffer produced by RF magnetron sputtering. Deposition on bare sapphire substrates resulted in a mixture of $α$-Ga$_2$O$_3$ and $ε$-Ga$_2$O$_3$ phases with a dislocation density of about $2\cdot10^{10}$ cm$^{-2}$. The insertion of $α$-Ga$_2$O$_3$ buffer layers resulted in phase-pure $α$-Ga$_2$O…
▽ More
Gallium oxide films were grown by HVPE on (0001) sapphire substrates with and without $α$-Cr$_2$O$_3$ buffer produced by RF magnetron sputtering. Deposition on bare sapphire substrates resulted in a mixture of $α$-Ga$_2$O$_3$ and $ε$-Ga$_2$O$_3$ phases with a dislocation density of about $2\cdot10^{10}$ cm$^{-2}$. The insertion of $α$-Ga$_2$O$_3$ buffer layers resulted in phase-pure $α$-Ga$_2$O$_3$ films and a fourfold reduction of the dislocation density to $5 \cdot 10^9$ cm$^{-2}$.
△ Less
Submitted 17 June, 2021;
originally announced June 2021.
-
Fano-type effect in hydrogen-terminated pure nanodiamond
Authors:
O. S. Kudryavtsev,
R. H. Bagramov,
A. M. Satanin,
A. A. Shiryaev,
O. I. Lebedev,
A. M. Romshin,
D. G. Pasternak,
A. V. Nikolaev,
V. P. Filonenko,
I. I. Vlasov
Abstract:
Two novel properties, unique for semiconductors: a negative electron affinity [1-2], and a high p-type surface electrical conductivity [3-4], were discovered in diamond at the end of the last century. Both properties appear when the diamond surface is hydrogenated. A natural question arises: is the influence of the surface hydrogen on diamond limited only to the electrical properties? Here, we rep…
▽ More
Two novel properties, unique for semiconductors: a negative electron affinity [1-2], and a high p-type surface electrical conductivity [3-4], were discovered in diamond at the end of the last century. Both properties appear when the diamond surface is hydrogenated. A natural question arises: is the influence of the surface hydrogen on diamond limited only to the electrical properties? Here, we report the first observation of a transparency peak at 1328 cm-1 in IR absorption of hydrogen-terminated pure (undoped) nanodiamonds. This new optical property is ascribed to Fano-type destructive interference between zone-center phonons and free carriers (holes) appearing in the near-surface layer of hydrogenated nanodiamond. Our work opens the way to exploring the physics of electron-phonon coupling in undoped diamonds and promises the application of the H-terminated nanodiamonds as a new optical material with an induced transparency in IR optical range.
△ Less
Submitted 6 June, 2021;
originally announced June 2021.
-
Phonon-assisted insulator-metal transitions in correlated systems driven by do**
Authors:
E. I. Shneyder,
M. V. Zotova,
S. V. Nikolaev,
S. G. Ovchinnikov
Abstract:
We consider how electron-phonon interaction influences the insulator-metal transitions driven by do** in the strongly correlated system. Using the polaronic version of the generalized tight-binding method, we investigate a multiband two-dimensional model taking into account both Holstein and Su-Schrieffer-Heeger types of electron-lattice contributions. For adiabatic ratio of the hop** paramete…
▽ More
We consider how electron-phonon interaction influences the insulator-metal transitions driven by do** in the strongly correlated system. Using the polaronic version of the generalized tight-binding method, we investigate a multiband two-dimensional model taking into account both Holstein and Su-Schrieffer-Heeger types of electron-lattice contributions. For adiabatic ratio of the hop** parameter and the phonon field energy, different types of band structure evolution are observed in a wide electron-phonon parameter range. We demonstrate the relationship between transition features and such properties of the system as the polaron and bipolaron crossovers, pseudogap behavior of various origin, orbital selectivity, and the redistribution of the spectral weight due to the electron-phonon interaction.
△ Less
Submitted 26 May, 2021;
originally announced May 2021.
-
Planning with Learned Entity Prompts for Abstractive Summarization
Authors:
Shashi Narayan,
Yao Zhao,
Joshua Maynez,
Gonçalo Simoes,
Vitaly Nikolaev,
Ryan McDonald
Abstract:
We introduce a simple but flexible mechanism to learn an intermediate plan to ground the generation of abstractive summaries. Specifically, we prepend (or prompt) target summaries with entity chains -- ordered sequences of entities mentioned in the summary. Transformer-based sequence-to-sequence models are then trained to generate the entity chain and then continue generating the summary condition…
▽ More
We introduce a simple but flexible mechanism to learn an intermediate plan to ground the generation of abstractive summaries. Specifically, we prepend (or prompt) target summaries with entity chains -- ordered sequences of entities mentioned in the summary. Transformer-based sequence-to-sequence models are then trained to generate the entity chain and then continue generating the summary conditioned on the entity chain and the input. We experimented with both pretraining and finetuning with this content planning objective. When evaluated on CNN/DailyMail, XSum, SAMSum and BillSum, we demonstrate empirically that the grounded generation with the planning objective improves entity specificity and planning in summaries for all datasets, and achieves state-of-the-art performance on XSum and SAMSum in terms of Rouge. Moreover, we demonstrate empirically that planning with entity chains provides a mechanism to control hallucinations in abstractive summaries. By prompting the decoder with a modified content plan that drops hallucinated entities, we outperform state-of-the-art approaches for faithfulness when evaluated automatically and by humans.
△ Less
Submitted 5 September, 2021; v1 submitted 15 April, 2021;
originally announced April 2021.
-
Saddle point anomaly of Landau levels in graphenelike structures
Authors:
A. V. Nikolaev
Abstract:
Studying the tight binding model in an applied rational magnetic field (H) we show that in graphene there are very unusual Landau levels situated in the immediate vicinity of the saddle point (M-point) energy epsilon_M. Landau levels around $ε_M$ are broadened into minibands (even in relatively weak magnetic fields ~40-53 T) with the maximal width reaching 0.4-0.5 of the energy separation between…
▽ More
Studying the tight binding model in an applied rational magnetic field (H) we show that in graphene there are very unusual Landau levels situated in the immediate vicinity of the saddle point (M-point) energy epsilon_M. Landau levels around $ε_M$ are broadened into minibands (even in relatively weak magnetic fields ~40-53 T) with the maximal width reaching 0.4-0.5 of the energy separation between two neighboring Landau levels though at all other energies the width of Landau levels is practically zero. In terms of the semiclassical approach a broad Landau level or magnetic miniband at epsilon_M is a manifestation of the so called self-intersecting orbit signifying an abrupt transition from the semiclassical trajectories enclosing the $Γ$ point to the trajectories enclosing the K point in the momentum space. Remarkably, the saddle point virtually does not affect the diamagnetic response of graphene, which is caused mostly by electron states in the vicinity of the Fermi energy ε_F. Experimentally, the effect of the broading of Landau levels can possibly be observed in twisted graphene where two saddle point singularities can be brought close to the Fermi energy.
△ Less
Submitted 22 December, 2021; v1 submitted 8 April, 2021;
originally announced April 2021.
-
Ion-Beam Modification of Metastable Gallium Oxide Polymorphs
Authors:
D. I. Tetelbaum,
A. A. Nikolskaya,
D. S. Korolev,
A. I. Belov,
V. N. Trushin,
Yu. A. Dudin,
A. N. Mikhaylov,
A. I. Pechnikov,
M. P. Scheglov,
V. I. Nikolaev,
D. Gogova
Abstract:
Gallium oxide with a corundum structure (α-Ga2O3) has recently attracted great attention in view of electronic and photonic applications due to its unique properties including a wide band gap exceeding that of the most stable beta phase (\b{eta}-Ga2O3). However, the lower thermal stability of the α-phase at ambient conditions in comparison with the \b{eta}-phase requires careful investigation of i…
▽ More
Gallium oxide with a corundum structure (α-Ga2O3) has recently attracted great attention in view of electronic and photonic applications due to its unique properties including a wide band gap exceeding that of the most stable beta phase (\b{eta}-Ga2O3). However, the lower thermal stability of the α-phase at ambient conditions in comparison with the \b{eta}-phase requires careful investigation of its resistance to other external influences such as ion irradiation, ion do**, etc. In this work, the structural changes under the action of Al+ ion irradiation have been investigated for a polymorphic gallium oxide layers grown by hydride vapor phase epitaxy on c-plane sapphire and consisting predominantly of α-phase with inclusions of α(\k{appa})-phase. It is established by the X-ray diffraction technique that inclusions of α(\k{appa})-phase in the irradiated layer undergo the expansion along the normal to the substrate surface, while there is no a noticeable deformation for the α-phase. This speaks in favor of the different radiation tolerance of various Ga2O3 polymorphs, especially the higher radiation tolerance of the α-phase. This fact should be taken into account when utilizing ion implantation to modify gallium oxide properties in terms of development of efficient do** strategies.
△ Less
Submitted 27 February, 2021;
originally announced March 2021.
-
The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics
Authors:
Sebastian Gehrmann,
Tosin Adewumi,
Karmanya Aggarwal,
Pawan Sasanka Ammanamanchi,
Aremu Anuoluwapo,
Antoine Bosselut,
Khyathi Raghavi Chandu,
Miruna Clinciu,
Dipanjan Das,
Kaustubh D. Dhole,
Wanyu Du,
Esin Durmus,
Ondřej Dušek,
Chris Emezue,
Varun Gangal,
Cristina Garbacea,
Tatsunori Hashimoto,
Yufang Hou,
Yacine Jernite,
Harsh Jhamtani,
Yangfeng Ji,
Shailza Jolly,
Mihir Kale,
Dhruv Kumar,
Faisal Ladhak
, et al. (31 additional authors not shown)
Abstract:
We introduce GEM, a living benchmark for natural language Generation (NLG), its Evaluation, and Metrics. Measuring progress in NLG relies on a constantly evolving ecosystem of automated metrics, datasets, and human evaluation standards. Due to this moving target, new models often still evaluate on divergent anglo-centric corpora with well-established, but flawed, metrics. This disconnect makes it…
▽ More
We introduce GEM, a living benchmark for natural language Generation (NLG), its Evaluation, and Metrics. Measuring progress in NLG relies on a constantly evolving ecosystem of automated metrics, datasets, and human evaluation standards. Due to this moving target, new models often still evaluate on divergent anglo-centric corpora with well-established, but flawed, metrics. This disconnect makes it challenging to identify the limitations of current models and opportunities for progress. Addressing this limitation, GEM provides an environment in which models can easily be applied to a wide set of tasks and in which evaluation strategies can be tested. Regular updates to the benchmark will help NLG research become more multilingual and evolve the challenge alongside models. This paper serves as the description of the data for which we are organizing a shared task at our ACL 2021 Workshop and to which we invite the entire NLG community to participate.
△ Less
Submitted 1 April, 2021; v1 submitted 2 February, 2021;
originally announced February 2021.
-
K-theory of Etesi C*-algebras
Authors:
Igor Nikolaev
Abstract:
We study the $C^*$-algebra $\mathbb{E}_{\mathscr{M}}$ of a smooth 4-dimensional manifold $\mathscr{M}$ introduced by Gábor Etesi. It is proved that the $\mathbb{E}_{\mathscr{M}}$ is a stationary AF-algebra. We calculate the topological and smooth invariants of $\mathscr{M}$ in terms of the K-theory of the $C^*$-algebra $\mathbb{E}_{\mathscr{M}}$. Using Gompf's Stable Diffeomorphism Theorem, it is…
▽ More
We study the $C^*$-algebra $\mathbb{E}_{\mathscr{M}}$ of a smooth 4-dimensional manifold $\mathscr{M}$ introduced by Gábor Etesi. It is proved that the $\mathbb{E}_{\mathscr{M}}$ is a stationary AF-algebra. We calculate the topological and smooth invariants of $\mathscr{M}$ in terms of the K-theory of the $C^*$-algebra $\mathbb{E}_{\mathscr{M}}$. Using Gompf's Stable Diffeomorphism Theorem, it is shown that all smoothings of $\mathscr{M}$ form a torsion abelian group. The latter is isomorphic to the Brauer group of a number field associated to the K-theory of $\mathbb{E}_{\mathscr{M}}$.
△ Less
Submitted 26 April, 2021; v1 submitted 15 December, 2020;
originally announced December 2020.
-
Ab initio based description of the unusual temperature increase of the electric field gradient at Ti sites in rutile TiO2
Authors:
A. V. Nikolaev,
N. M. Chtchelkatchev,
A. V. Bibikov,
D. A. Salamatin,
A. V. Tsvyashchenko
Abstract:
Combining a precise ab initio electron band structure calculation of the TiO2 rutile structure with the temperature evolution of the Ti mean-square displacements, we reproduce a puzzling temperature increase of the electric field gradient at Ti sites in TiO2, observed experimentally. Our method employs a procedure of averaging two quadrupole electron density components (L = 2) inside a sphere vibr…
▽ More
Combining a precise ab initio electron band structure calculation of the TiO2 rutile structure with the temperature evolution of the Ti mean-square displacements, we reproduce a puzzling temperature increase of the electric field gradient at Ti sites in TiO2, observed experimentally. Our method employs a procedure of averaging two quadrupole electron density components (L = 2) inside a sphere vibrating with the Ti nucleus at its center, where the key factor introducing the temperature dependence is the square root of the Debye-Waller factor. Although the Debye-Waller factor always reduces the corresponding Fourier component, in TiO2 due to the interplay between terms of opposite signs, it results in a net increase of the whole sum with temperature, leading to the growth of the electric field gradient. Quantitatively, we find that the increase of electric field gradient is only half of the experimental value, which we ascribe to anharmonic effects or a strong oxygen position influence. In addition, our method reproduces the unusual temperature dependence of the asymmetry parameter eta, which first decreases with temperature, goes to zero and then increases.
△ Less
Submitted 11 November, 2020;
originally announced November 2020.
-
Convolution Neural Networks for Semantic Segmentation: Application to Small Datasets of Biomedical Images
Authors:
Vitaly Nikolaev
Abstract:
This thesis studies how the segmentation results, produced by convolutional neural networks (CNN), is different from each other when applied to small biomedical datasets. We use different architectures, parameters and hyper-parameters, trying to find out the better configurations for our task, and trying to find out underlying regularities. Two working datasets are from biomedical area of research…
▽ More
This thesis studies how the segmentation results, produced by convolutional neural networks (CNN), is different from each other when applied to small biomedical datasets. We use different architectures, parameters and hyper-parameters, trying to find out the better configurations for our task, and trying to find out underlying regularities. Two working datasets are from biomedical area of research. We conducted a lot of experiments with the two types of networks and the received results have shown the preference of some conditions of experiments and parameters of the networks over the others. All testing results are given in the tables and some selected resulting graphs and segmentation predictions are shown for better illustration.
△ Less
Submitted 1 November, 2020;
originally announced November 2020.
-
Peculiar chemical bonding between thorium and a carbon hexagon in carbon nanomaterials
Authors:
A. V. Bibikov,
A. V. Nikolaev,
E. V. Tkalya
Abstract:
We explore an unusual nature of chemical bonding of the thorium atom with a ring of six carbon atoms (hexagon) in novel carbon materials. Our ab initio calculations of Th-based metallofullerenes (Th@C60, Th@C20) and Th bound to benzene or coronene at the Hartree-Fock level with the second order perturbation (MP2) correction accounting for the van der Waals interactions, demonstrate that the optima…
▽ More
We explore an unusual nature of chemical bonding of the thorium atom with a ring of six carbon atoms (hexagon) in novel carbon materials. Our ab initio calculations of Th-based metallofullerenes (Th@C60, Th@C20) and Th bound to benzene or coronene at the Hartree-Fock level with the second order perturbation (MP2) correction accounting for the van der Waals interactions, demonstrate that the optimal position of the thorium atom is where it faces the center of a hexagon and is located at a distance of 2.01-2.07 A from the center. For Th encapsulated in C60 it is found at 2.01 A, whereas the other local energy minima are shifted to larger energies (0.22 eV and higher). Inside C60 the highest local minimum at 1.17 eV is observed when Th faces the center of the five member carbon ring (pentagon). Based on our calculations for Th with benzene and coronene where the global minimum for Th corresponds to its position at 2.05 A (benzene) or 2.02 A (coronene) from the hexagon center, we conclude that a well pronounced minimum is likely to present in graphene and in a single wall carbon nanotube. The ground state of Th is singlet, other high spin states (triplet and quintet) lie higher in energy (> 1.62 eV). We discuss a potential use of the carbon nanomaterials with the 229Th isotope having the nuclear transition of the optical range, for metrological purposes.
△ Less
Submitted 16 September, 2020;
originally announced September 2020.
-
Backtracking algorithms for constructing the Hamiltonian decomposition of a 4-regular multigraph
Authors:
Alexander V. Korostil,
Andrei V. Nikolaev
Abstract:
We consider a Hamiltonian decomposition problem of partitioning a regular graph into edge-disjoint Hamiltonian cycles. It is known that verifying vertex non-adjacency in the 1-skeleton of the symmetric and asymmetric traveling salesperson polytopes is NP-complete. On the other hand, a sufficient condition for two vertices to be non-adjacent can be formulated as a combinatorial problem of finding a…
▽ More
We consider a Hamiltonian decomposition problem of partitioning a regular graph into edge-disjoint Hamiltonian cycles. It is known that verifying vertex non-adjacency in the 1-skeleton of the symmetric and asymmetric traveling salesperson polytopes is NP-complete. On the other hand, a sufficient condition for two vertices to be non-adjacent can be formulated as a combinatorial problem of finding a second Hamiltonian decomposition of a 4-regular multigraph. We present two backtracking algorithms for constructing a second Hamiltonian decomposition and verifying vertex non-adjacency: an algorithm based on a simple path extension and an algorithm based on the chain edge fixing procedure.
Based on the results of computational experiments for undirected multigraphs, both backtracking algorithms lost to the known general variable neighborhood search heuristics. However, for directed multigraphs, the algorithm based on chain fixing of edges showed results comparable to heuristics on instances with an existing solution and better results on infeasible instances where the Hamiltonian decomposition does not exist.
△ Less
Submitted 26 May, 2022; v1 submitted 10 September, 2020;
originally announced September 2020.
-
Remark on Faltings theorem
Authors:
Igor Nikolaev
Abstract:
We prove Faltings Finiteness Theorem using Rieffel's classification of the noncommutative tori.
We prove Faltings Finiteness Theorem using Rieffel's classification of the noncommutative tori.
△ Less
Submitted 18 August, 2021; v1 submitted 2 September, 2020;
originally announced September 2020.
