Search

James Webb Space Telescope (JWST) has discovered many faint AGNs at high-$z$ by detecting their broad Balmer lines. However, their high number density, lack of X-ray emission, and overly high black hole masses with respect to their host stellar masses suggest that they are a distinct population from general type-1 quasars. Here, we present clustering analysis of 28 low-luminosity broad-line AGNs found by JWST (JWST AGNs) at $5

The structure of an Al(111)3x3-Si surface was examined by combining data from positron diffraction and core-level photoemission spectroscopy. Analysis of the diffraction rocking curves indicated that the overlayer had a flat honeycomb lattice structure. Simulations of Si core-level spectra calculated via the first-principles indicated that one of the Si atoms in the unit cell was replaced by an Al atom. The surface superstructure was thus a two-dimensional layer of Al-embedded silicene on Al(111).

In this paper, we rethink sparse lexical representations for image retrieval. By utilizing multi-modal large language models (M-LLMs) that support visual prompting, we can extract image features and convert them into textual data, enabling us to utilize efficient sparse retrieval algorithms employed in natural language processing for image retrieval tasks. To assist the LLM in extracting image features, we apply data augmentation techniques for key expansion and analyze the impact with a metric for relevance between images and textual data. We empirically show the superior precision and recall performance of our image retrieval method compared to conventional vision-language model-based methods on the MS-COCO, PASCAL VOC, and NUS-WIDE datasets in a keyword-based image retrieval scenario, where keywords serve as search queries. We also demonstrate that the retrieval performance can be improved by iteratively incorporating keywords into search queries., Comment: Accepted to ECCV 2024 Workshops: 2nd Workshop on Traditional Computer Vision in the Age of Deep Learning (TradiCV)

In quantum information processing, two primary research directions have emerged: one based on discrete variables (DV) and the other on the structure of quantum states in a continuous-variable (CV) space. It is increasingly recognized that integrating these two approaches could unlock new potentials, overcoming the inherent limitations of each. Here, we show that such a DV-CV hybrid approach, applied to superconducting Kerr parametric oscillators (KPOs), enables us to entangle a pair of Schr\"odinger's cat states by two straightforward methods. The first method involves the entanglement-preserving and deterministic conversion between Bell states in the Fock-state basis (DV encoding) and those in the cat-state basis (CV encoding). This method would allow us to construct quantum networks in the cat-state basis using conventional schemes originally developed for the Fock-state basis. In the second method, the $\sqrt{\textrm{iSWAP}}$ gate operation is implemented between two cat states following the procedure used for Fock-state encoding. This DV-like gate operation on CV encoding not only completes the demonstration of a universal quantum gate set in a KPO system but also enables faster and simpler gate operations compared to previous SWAP gate implementations on bosonic modes. Our work offers a simple yet powerful application of DV-CV hybridization while also highlighting the scalability of this planar KPO system.

We present the study on the relationship between SMBHs and their host galaxies using our variability-selected AGN sample ($i_\mathrm{AB} \leq 25.9,\ z \leq 4.5$) constructed from the HSC-SSP Ultra-deep survey in the COSMOS field. We estimated the BH mass ($M_\mathrm{BH}=10^{5.5-10}\ M_{\odot}$) based on the single-epoch virial method and the total stellar mass ($M_\mathrm{star}=10^{10-12}\ M_{\odot}$) by separating the AGN component with SED fitting. We found that the redshift evolution of the BH-stellar mass ratio ($M_\mathrm{BH}/M_\mathrm{star}$) depends on the $M_\mathrm{BH}$ which is caused by the no significant correlation between $M_\mathrm{BH}$ and $M_\mathrm{star}$. Variable AGNs with massive SMBHs ($M_\mathrm{BH}>10^{9}\ M_{\odot}$) at $1.5\sim1\%$) than the BH-bulge ratios ($M_\mathrm{BH}/M_\mathrm{bulge}$) observed in the local universe for the same BH range. This implies that there is a typical growth path of massive SMBHs which is faster than the formation of the bulge component as final products seen in the present day. For the low-mass SMBHs ($M_\mathrm{BH}<10^{8}\ M_{\odot}$) at $0.5 4$. We interpret that host galaxies harboring less massive SMBHs at intermediate redshift have already acquired sufficient stellar mass, although high-z galaxies are still in the early stage of galaxy formation relative to those at the intermediate/local universe., Comment: Accepted for publication in ApJ, 14 pages, 5 figures

Let $k$ be a field, $T$ be an algebraic $k$-torus, $X$ be a smooth $k$-compactification of $T$ and ${\rm Pic}\,\overline{X}$ be the Picard group of $\overline{X}=X\times_k\overline{k}$ where $\overline{k}$ is a fixed separable closure of $k$. Hoshi, Kanai and Yamasaki [HKY22], [HKY23] determined $H^1(k,{\rm Pic}\, \overline{X})$ for norm one tori $T=R^{(1)}_{K/k}(\mathbb{G}_m)$ and gave a necessary and sufficient condition for the Hasse norm principle for extensions $K/k$ of number fields with $[K:k]\leq 15$. In this paper, we treat the case where $[K:k]=16$. Among $1954$ transitive subgroups $G=16Tm\leq S_{16}$ $(1\leq m\leq 1954)$ up to conjugacy, we determine $1101$ (resp. $774$, $31$, $37$, $1$, $1$, $9$) cases with $H^1(k,{\rm Pic}\, \overline{X})=0$ (resp. $Z/2Z$, $(Z/2Z)^{\oplus 2}$, $(Z/2Z)^{\oplus 3}$, $(Z/2Z)^{\oplus 4}$, $(Z/2Z)^{\oplus 6}$, $Z/4Z$) where $G$ is the Galois group of the Galois closure $L/k$ of $K/k$. We see that $H^1(k,{\rm Pic}\, \overline{X})=0$ implies that the Hasse norm principle holds for $K/k$. In particular, among $22$ primitive $G=16Tm$ cases, i.e. $H\leq G=16Tm$ is maximal with $[G:H]=16$, we determine exactly $6$ cases $(m=178, 708, 1080, 1329, 1654, 1753)$ with $H^1(k,{\rm Pic}\, \overline{X})\neq 0$ $($$(Z/2Z)^{\oplus 2}$, $Z/2Z$, $(Z/2Z)^{\oplus 2}$, $Z/2Z$, $Z/2Z$, $Z/2Z$). Moreover, we give a necessary and sufficient condition for the Hasse norm principle for $K/k$ with $[K:k]=16$ for $22$ primitive $G=16Tm$ cases. As a consequence of the $22$ primitive $G$ cases, we get the Tamagawa number $\tau(T)=1$, $1/2$, $1/4$ of $T=R^{(1)}_{K/k}(\mathbb{G}_m)$ over a number field $k$ via Ono's formula $\tau(T)=1/|Sha(T)|$ where $Sha(T)$ is the Shafarevich-Tate group of $T$., Comment: 74 pages. arXiv admin note: text overlap with arXiv:2210.09119

SDSS J1430+2303 has been argued to possess a supermassive black hole binary which is predicted to merge within a few months or three years from January 2022. We conducted follow-up optical spectroscopic observations of SDSS J1430+2303 with KOOLS-IFU on Seimei Telescope in May, June, and July 2022, and April 2023. The observed spectrum around $\mathrm{H}\mathrm{\alpha}$ shows a central broad component $\sim 10^3\ \mathrm{km\ s^{-1}}$ blueshifted from the narrow H$\mathrm{\alpha}$ line as well as the broader double-peaked component with a separation of $\sim\pm 5\times10^3\ \mathrm{km\ s^{-1}}$, similar to the spectrum reported in January 2022. We investigate the variability of the complex broad $\mathrm{H}\mathrm{\alpha}$ emission line relative to the continuum over the observation period. The continuum-normalized relative flux of the central broad component shows the increasing trend from May to July 2022 which is interpreted to be caused by the decrease of the continuum as also supported by damping of the X-ray, UV, and optical light curves observed for the same period. From July 2022 to April 2023, however, the central broad component decreased significantly. For the relative flux of the broader double-peaked component, on the other hand, no significant change appears at any epoch. These results suggest that the complicated broad line profile of SDSS J1430+2303 is generated from at least two distinct regions. While the central broad component originates from a broad line region, the broader double-peaked component arises in the vicinity of the continuum source., Comment: Published in Publications of the Astronomical Society of Japan. 6 pages, 3 figures

We give a complete answer to the rationality problem (up to stable $k$-equivalence) for norm one tori $T=R^{(1)}_{K/k}(\mathbb{G}_m)$ of $K/k$ whose Galois closures $L/k$ are $A_5\simeq {\rm PSL}_2(\mathbb{F}_4)$ and ${\rm PSL}_2(\mathbb{F}_8)$ extensions. In particular, we prove that $T$ is stably $k$-rational for $G={\rm Gal}(L/k)\simeq {\rm PSL}_2(\mathbb{F}_{8})$, $H={\rm Gal}(L/K)\simeq (C_2)^3$ and $H\simeq (C_2)^3\rtimes C_7$ where $C_n$ is the cyclic group of order $n$. Based on the result, we conjecture that $T$ is stably $k$-rational for $G\simeq {\rm PSL}_2(\mathbb{F}_{2^d})$, $H\simeq (C_2)^d$ and $H\simeq (C_2)^d\rtimes C_{2^d-1}$. Some other cases $G\simeq A_n$, $S_n$, ${\rm GL}_n(\mathbb{F}_{p^d})$, ${\rm SL}_n(\mathbb{F}_{p^d})$, ${\rm PGL}_n(\mathbb{F}_{p^d})$, ${\rm PSL}_n(\mathbb{F}_{p^d})$ and $H\lneq G$ are also investigated for small $n$ and $p^d$., Comment: 26 pages. Data files PSL28E8bpS.dat and PSL28bpQ.dat are added to the second-named author's web page. arXiv admin note: text overlap with arXiv:2302.06231, arXiv:1910.01469, arXiv:1811.02145, arXiv:1210.4525

We report a systematic quantitative evaluation of parametric amplification gain of magnetization dynamics in ytirrium iron garnet ($\mathrm{Y}_3\mathrm{Fe}_5\mathrm{O}_{12}$) thin disk via a.c. spin pumping and inverse spin Hall effect. We demonstrate its signature phase-dependence where amplification and attenuation occur every $\frac{\pi}{2}$ phase shift of the input signal. The results also show the pump-power dependence of the gain that is explained well by our theoretical model. Finally, the optimal conditions for the amplification is investigated by measuring the magnetic field dependence, where we find the highest gain of 11.4 dB., Comment: 6 pages, 3 figures, 14 panels

This study investigates engagement patterns related to OpenAI's ChatGPT on Japanese Twitter, focusing on two distinct user groups - early and late engagers, inspired by the Innovation Theory. Early engagers are defined as individuals who initiated conversations about ChatGPT during its early stages, whereas late engagers are those who began participating at a later date. To examine the nature of the conversations, we employ a dual methodology, encompassing both quantitative and qualitative analyses. The quantitative analysis reveals that early engagers often engage with more forward-looking and speculative topics, emphasizing the technological advancements and potential transformative impact of ChatGPT. Conversely, the late engagers intereact more with contemporary topics, focusing on the optimization of existing AI capabilities and considering their inherent limitations. Through our qualitative analysis, we propose a method to measure the proportion of shared or unique viewpoints within topics across both groups. We found that early engagers generally concentrate on a more limited range of perspectives, whereas late engagers exhibit a wider range of viewpoints. Interestingly, a weak correlation was found between the volume of tweets and the diversity of discussed topics in both groups. These findings underscore the importance of identifying semantic bias, rather than relying solely on the volume of tweets, for understanding differences in communication styles between groups within a given topic. Moreover, our versatile dual methodology holds potential for broader applications, such as studying engagement patterns within different user groups, or in contexts beyond ChatGPT.

Retrieval-augmented large language models (R-LLMs) combine pre-trained large language models (LLMs) with information retrieval systems to improve the accuracy of factual question-answering. However, current libraries for building R-LLMs provide high-level abstractions without sufficient transparency for evaluating and optimizing prompts within specific inference processes such as retrieval and generation. To address this gap, we present RaLLe, an open-source framework designed to facilitate the development, evaluation, and optimization of R-LLMs for knowledge-intensive tasks. With RaLLe, developers can easily develop and evaluate R-LLMs, improving hand-crafted prompts, assessing individual inference processes, and objectively measuring overall system performance quantitatively. By leveraging these features, developers can enhance the performance and accuracy of their R-LLMs in knowledge-intensive generation tasks. We open-source our code at https://github.com/yhoshi3/RaLLe., Comment: 18 pages, 2 figures, see https://youtu.be/JYbm75qnfTg for the demonstration screencast, accepted by EMNLP 2023 System Demonstrations

Recently, robustness of superconductivity (transition temperature, Tc) under high pressures has been observed in high-entropy alloy (HEA), bcc-type Ti-Zr-Hf-Nb-Ta, and HEA-type compounds (Ag,In,Sn,Pb,Bi)Te with a NaCl-type structure. Since those materials have three-dimensional crystal structure, investigation on the pressure dependence of Tc of low-dimensional materials is needed to understand the phenomena. Here, we investigated the superconducting properties and the crystal structure of BiS2-based layered system REO0.5F0.5BiS2. Although the robustness of Tc was induced in MTe with increasing M-site configurational entropy, the increase in RE-site configurational entropy does not induce robustness of Tc under high pressures in REO0.5F0.5BiS2. The crystal structure of HEA-type REO0.5F0.5BiS2 was confirmed as monoclinic P21/m, which is the same space group as the zero-entropy counterpart LaO0.5F0.5BiS2. The results suggest that an increase in configurational entropy at blocking layers do not affect crystal structure and superconducting properties under high pressures in the BiS2-based layered system., Comment: 10 pages, 4 figures, to appear in JPSJ

Large Language Model (LLM) based Generative AI systems have seen significant progress in recent years. Integrating a knowledge retrieval architecture allows for seamless integration of private data into publicly available Generative AI systems using pre-trained LLM without requiring additional model fine-tuning. Moreover, Retrieval-Centric Generation (RCG) approach, a promising future research direction that explicitly separates roles of LLMs and retrievers in context interpretation and knowledge memorization, potentially leads to more efficient implementation. SimplyRetrieve is an open-source tool with the goal of providing a localized, lightweight, and user-friendly interface to these sophisticated advancements to the machine learning community. SimplyRetrieve features a GUI and API based RCG platform, assisted by a Private Knowledge Base Constructor and a Retrieval Tuning Module. By leveraging these capabilities, users can explore the potential of RCG for improving generative AI performance while maintaining privacy standards. The tool is available at https://github.com/RCGAI/SimplyRetrieve with an MIT license., Comment: 12 pages, 6 figures

We explored quinary body-centered cubic (bcc) high-entropy alloy (HEA) superconductors with valence electron concentrations (VECs) ranging from 4.6 to 5.0, a domain that has received limited attention in prior research. Our search has led to the discovery of new bcc Ti-Hf-Nb-Ta-Re superconducting alloys, which exhibit an interesting phenomenon of phase segregation into two bcc phases with slightly different chemical compositions, as the VEC increases. The enthalpy of the formation of each binary compound explains the phase segregation. All the alloys investigated were categorized as type-II superconductors, with superconducting critical temperatures ($T_\mathrm{c}$) ranging from 3.25 K to 4.38 K. We measured the Vickers microhardness, which positively correlated with the Debye temperature, and compared it with the hardness values of other bcc HEA superconductors. Our results indicate that $T_\mathrm{c}$ systematically decreases with an increase in hardness beyond a threshold of approximately 350 HV. Additionally, we plotted $T_\mathrm{c}$ vs. VEC for representative quinary bcc HEAs. The plot revealed the asymmetric VEC dependence. The correlation between the hardness and $T_\mathrm{c}$, as well as the asymmetric dependence of $T_\mathrm{c}$ on VEC can be attributed to the simultaneous effects of the electronic density of states at the Fermi level and electron-phonon coupling under the uncertainty principle, especially in the higher VEC region., Comment: to appear in Journal of Alloys and Metallurgical Systems

We constructed a large-scale electromagnetic calorimeter prototype as a part of the Forward Calorimeter upgrade project (FoCal) for the ALICE experiment at the Large Hadron Collider (LHC). The prototype, also known as ``Mini FoCal'', consists of 20 layers of silicon pad sensors and tungsten alloy plates with printed circuit boards and readout electronics. The constructed detector was tested at the test beam facility of the Super Proton Synchrotron (SPS) at CERN. We obtain an energy resolution of about 4.3% for electron beams at both 150 and 250 GeV/$c$, which is consistent with realistic detector response simulations. Longitudinal profiles of electromagnetic shower were also measured and found to agree with the simulations. The same prototype detector was installed in the ALICE experimental area about 7.5m away from the interaction point. It was used to measure inclusive electromagnetic cluster energy distributions and neutral-pion candidate invariant mass distributions for pseudo-rapidity of $\eta$=3.7-4.5 in proton-proton collisions at $\sqrt{s}$ = 13 TeV at LHC. The measured distributions in different $\eta$ regions are similar to those obtained from PYTHIA simulations., Comment: 30 pages, 27 figures

We present a new computation method for simulating reflection high-energy electron diffraction and the total-reflection high-energy positron diffraction experiments. The two experiments are used commonly for the structural analysis of material surface. The present paper improves the conventional numerical method, the multi-slice method, for faster computation, since the present method avoids the matrix-eigenvalue solver for the computation of matrix exponentials and can adopt higher-order ordinary differential equation solvers. Moreover, we propose a high-performance implementation based on multi-thread parallelization and cache-reusable subroutines. In our tests, this new method performs up to 2,000 times faster than the conventional method.

Parametric oscillation occurs when the resonance frequency of an oscillator is periodically modulated. Owing to time-reversal symmetry breaking in magnets, nonreciprocal magnons can be parametrically excited when spatial-inversion symmetry breaking is provided. This means that magnons with opposite propagation directions have different amplitudes. Here we demonstrate switching on and off the magnon parametric oscillation by reversing the external field direction applied to a Y$_3$Fe$_5$O$_{12}$ micro-structured film. The result originates from the nonreciprocity of surface mode magnons, leading to field-direction dependence of the magnon accumulation under a nonuniform microwave pumping. Our numerical calculation well reproduces the experimental result., Comment: 5 pages, 4 figures

The identity by Chaundy and Bullard expresses $1$ as a sum of two truncated binomial series in one variable where the truncations depend on two different non-negative integers. We present basic and elliptic extensions of the Chaundy--Bullard identity. The most general result, the elliptic extension, involves, in addition to the nome $p$ and the base $q$, four independent complex variables. Our proof uses a suitable weighted lattice path model. We also show how three of the basic extensions can be viewed as B\'ezout identities. Inspired by the lattice path model, we give a new elliptic extension of the binomial theorem, taking the form of an identity for elliptic commuting variables. We further present variants of the homogeneous form of the identity for $q$-commuting and for elliptic commuting variables., Comment: 22 pages, dedicated to the memory of Richard Allen Askey; minor changes

Spin Hall effect of spin-texture origin is explored theoretically for antiferromagnetic (AF) metals. It is found that a vector chirality formed by the N\'eel vector gives rise to a topological spin Hall effect. This is topological since it is proportional to the winding number counted by in-plane vector chirality along the sample edge, which can be nonvanishing for AF merons but not for AF skyrmions. The effect is enhanced when the Fermi level lies near the AF gap, and, surprisingly, at weak coupling with small AF gap. These features are confirmed numerically based on the Landauer-B\"uttiker formula. Important roles played by nonadiabatic processes and spin dephasing are pointed out., Comment: 16 pages, 11 figures

Neural document retrievers, including dense passage retrieval (DPR), have outperformed classical lexical-matching retrievers, such as BM25, when fine-tuned and tested on specific question-answering datasets. However, it has been shown that the existing dense retrievers do not generalize well not only out of domain but even in domain such as Wikipedia, especially when a named entity in a question is a dominant clue for retrieval. In this paper, we propose an approach toward in-domain generalization using the embeddings generated by the frozen language model trained with the entities in the domain. By not fine-tuning, we explore the possibility that the rich knowledge contained in a pretrained language model can be used for retrieval tasks. The proposed method outperforms conventional DPRs on entity-centric questions in Wikipedia domain and achieves almost comparable performance to BM25 and state-of-the-art SPAR model. We also show that the contextualized keys lead to strong improvements compared to BM25 when the entity names consist of common words. Our results demonstrate the feasibility of the zero-shot retrieval method for entity-centric questions of Wikipedia domain, where DPR has struggled to perform., Comment: Accepted to Workshop on Knowledge Augmented Methods for Natural Language Processing, in conjunction with AAAI 2023