Search

We analyze the line defect correlation functions decorating the supersymmetric indices of $\mathcal{N}=4$ super Yang-Mills theories with orthogonal and symplectic gauge groups. We obtain exact closed-form expressions for the correlators by means of combinatorial methods associated with the group characters, including the generalized Murnaghan-Nakayama rules. The exact results for the large $N$ correlators lead to the gravity indices for the gravity dual spectrum of the fluctuations of the fundamental string wrapping $AdS_2$ and those of the fat string, the D5-brane wrapping $AdS_2$ $\times$ $\mathbb{RP}^4$., Comment: 88 pages

In recent neutrino detectors, neutrons produced in neutrino reactions play an important role. Muon capture on oxygen nuclei is one of the processes that produce neutrons in water Cherenkov detectors. We measured neutron multiplicity in the process using cosmic ray muons that stop in the gadolinium-loaded Super-Kamiokande detector. For this measurement, neutron detection efficiency is obtained with the muon capture events followed by gamma rays to be $50.2^{+2.0}_{-2.1}\%$. By fitting the observed multiplicity considering the detection efficiency, we measure neutron multiplicity in muon capture as $P(0)=24\pm3\%$, $P(1)=70^{+3}_{-2}\%$, $P(2)=6.1\pm0.5\%$, $P(3)=0.38\pm0.09\%$. This is the first measurement of the multiplicity of neutrons associated with muon capture without neutron energy threshold.

Binarity in massive stars has proven to be an important aspect in the their evolution. For Be stars, it might be the cause of their spin up, and thus part of the mechanism behind the formation of their viscous decretion disks. Detecting companions in systems with Be stars is challenging, making it difficult to obtain observational constraints on their binary fraction. We explore the effects of a binary companion in a system with a Be star, from disk formation to quasi steady-state using smoothed particle hydrodynamics (SPH) simulations of coplanar, circular binary systems. High spatial resolution is achieved by adopting particle splitting in the SPH code, as well as a more realistic description of the secondary star and the disk viscosity. The tidal forces considerably affect the Be disk, forming distinct regions in the system, with observational consequences that can be used to infer the presence of a otherwise undetectable companion. With the upgraded code, we can probe a region approximately 4 times larger than previously possible. We describe the configuration and kinematics of each part of the system, and provide a summary of their expected observational signals. Material that enters the Roche lobe of the companion is partially captured by it, forming a rotationally supported, disk-like structure. Material not accreted escapes and forms a circumbinary disk around the system. This is the first work to describe the region beyond the truncation region of the Be disk and its observational consequences with detail. We argue that observational features of previously unclear origin, such as the intermittent shell features and emission features of HR 2142 and HD 55606, originate in areas beyond the truncation region. This new understanding of the behavior of disks in Be binaries will allow not just for better interpretation of existing data, but also for the planning of future observations.

Detecting texts generated by Large Language Models (LLMs) could cause grave mistakes due to incorrect decisions, such as undermining student's academic dignity. LLM text detection thus needs to ensure the interpretability of the decision, which can help users judge how reliably correct its prediction is. When humans verify whether a text is human-written or LLM-generated, they intuitively investigate with which of them it shares more similar spans. However, existing interpretable detectors are not aligned with the human decision-making process and fail to offer evidence that users easily understand. To bridge this gap, we introduce ExaGPT, an interpretable detection approach grounded in the human decision-making process for verifying the origin of a text. ExaGPT identifies a text by checking whether it shares more similar spans with human-written vs. with LLM-generated texts from a datastore. This approach can provide similar span examples that contribute to the decision for each span in the text as evidence. Our human evaluation demonstrates that providing similar span examples contributes more effectively to judging the correctness of the decision than existing interpretable methods. Moreover, extensive experiments in four domains and three generators show that ExaGPT massively outperforms prior powerful detectors by up to +40.9 points of accuracy at a false positive rate of 1%., Comment: Under Review

In this study, we propose Explainable Multimodal Machine Learning (EMML), which integrates the analysis of diverse data types (multimodal data) using factor analysis for feature extraction with Explainable AI (XAI), for carbon nanotube (CNT) fibers prepared from aqueous dispersions. This method is a powerful approach to elucidate the mechanisms governing material properties, where multi-stage fabrication conditions and multiscale structures have complex influences. Thus, in our case, this approach helps us understand how different processing steps and structures at various scales impact the final properties of CNT fibers. The analysis targeted structures ranging from the nanoscale to the macroscale, including aggregation size distributions of CNT dispersions and the effective length of CNTs. Furthermore, because some types of data were difficult to interpret using standard methods, challenging-to-interpret distribution data were analyzed using Negative Matrix Factorization (NMF) for extracting key features that determine the outcome. Contribution analysis with SHapley Additive exPlanations (SHAP) demonstrated that small, uniformly distributed aggregates are crucial for improving fracture strength, while CNTs with long effective lengths are significant factors for enhancing electrical conductivity. The analysis also identified thresholds and trends for these key factors to assist in defining the conditions needed to optimize CNT fiber properties. EMML is not limited to CNT fibers but can be applied to the design of other materials derived from nanomaterials, making it a useful tool for developing a wide range of advanced materials. This approach provides a foundation for advancing data-driven materials research., Comment: 33 pages, 9 figures

The rise of large language models (LLMs) has led to more diverse and higher-quality machine-generated text. However, their high expressive power makes it difficult to control outputs based on specific business instructions. In response, benchmarks focusing on the controllability of LLMs have been developed, but several issues remain: (1) They primarily cover major languages like English and Chinese, neglecting low-resource languages like Japanese; (2) Current benchmarks employ task-specific evaluation metrics, lacking a unified framework for selecting models based on controllability across different use cases. To address these challenges, this research introduces LCTG Bench, the first Japanese benchmark for evaluating the controllability of LLMs. LCTG Bench provides a unified framework for assessing control performance, enabling users to select the most suitable model for their use cases based on controllability. By evaluating nine diverse Japanese-specific and multilingual LLMs like GPT-4, we highlight the current state and challenges of controllability in Japanese LLMs and reveal the significant gap between multilingual models and Japanese-specific models., Comment: 15 pages, 11 figures. Project page: this [URL](https://github.com/CyberAgentAILab/LCTG-Bench)

Vision-language models (VLMs) have shown impressive abilities across a range of multi-modal tasks. However, existing metrics for evaluating the quality of text generated by VLMs typically focus on an overall evaluation for a specific task, such as image captioning. While the overall evaluation is essential for any task, the criteria prioritized can differ depending on the task, making it challenging for current metrics to adapt to multi-task scenarios. To address this limitation, we propose HarmonicEval, a reference-free comprehensive evaluation metric that aggregates criterion-wise scores to produce the overall score in a bottom-up manner. Furthermore, we construct the Multi-task Multi-criteria Human Evaluation (MMHE) dataset, which comprises 18,000 expert human judgments across four multi-modal tasks. Our experiments demonstrate that HarmonicEval achieves higher correlations with human judgments than conventional metrics while providing numerical scores for each criterion.

Why do we build local large language models (LLMs)? What should a local LLM learn from the target language? Which abilities can be transferred from other languages? Do language-specific scaling laws exist? To explore these research questions, we evaluated 35 Japanese, English, and multilingual LLMs on 19 evaluation benchmarks for Japanese and English, taking Japanese as a local language. Adopting an observational approach, we analyzed correlations of benchmark scores, and conducted principal component analysis (PCA) on the scores to derive \textit{ability factors} of local LLMs. We found that training on English text can improve the scores of academic subjects in Japanese (JMMLU). In addition, it is unnecessary to specifically train on Japanese text to enhance abilities for solving Japanese code generation, arithmetic reasoning, commonsense, and reading comprehension tasks. In contrast, training on Japanese text could improve question-answering tasks about Japanese knowledge and English-Japanese translation, which indicates that abilities for solving these two tasks can be regarded as \textit{Japanese abilities} for LLMs. Furthermore, we confirmed that the Japanese abilities scale with the computational budget for Japanese text., Comment: Preprint. Under review

Non-equilibrium states generated by ultrafast laser pulses are characterized by specific phenomena that are not accessible in static measurements. Previous time- and angle-resolved photoemission spectroscopy (TARPES) studies on rare-earth tritelluride materials have revealed the laser-driven melting of the charge density wave order as well as its collective amplitude mode excitation. Variation of the excess energy deposited by optical pumping in the material promises pathways to control the dynamic material response. To this end, we use an optical parametric amplifier to generate a tunable pump photon energy. Studying LaTe3 we compare the dynamics driven by pumping resonantly across the charge density wave energy gap with the effect of pumping at a twice higher photon energy in a TARPES pump-probe experiment. We clearly identify a pump photon energy dependent behavior. At the larger pump photon energy, the excess electronic energy generates lattice heating mediated by e-ph coupling and softening of the amplitude mode frequency from 3 to 2 THz. Remarkably, the resonant pumping across the CDW gap results in a time-independent amplitude mode frequency. We conclude that the resonant excitation across the energy gap excites the amplitude mode selectively while additional electronic excess energy deposited at higher pump photon energy modifies the crystal properties transiently by incoherent dissipative processes., Comment: 10 pages, 3 figures

The experimental realisation of unconventional superconductivity and charge order in kagome systems \textit{A}V$_3$Sb$_5$ is of critical importance. We conducted a highly systematic study of Cs(V$_{1-x}$Nb$_x$)$_3$Sb$_5$ with $x$=0.07 (Nb$_{0.07}$-CVS) by employing a unique combination of tuning parameters such as doping, hydrostatic pressure, magnetic fields, and depth, using muon spin rotation, AC susceptibility, and STM. We uncovered tunable magnetism in the normal state of Nb$_{0.07}$-CVS, which transitions to a time-reversal symmetry (TRS) breaking superconducting state under pressure. Specifically, our findings reveal that the bulk of Nb$_{0.07}$-CVS (at depths greater than 20 nm from the surface) experiences TRS breaking below $T^*=40~$K, lower than the charge order onset temperature, $T_\mathrm{CO}$ = 58 K. However, near the surface (within 20 nm from the surface), the TRS breaking signal doubles and onsets at $T_\mathrm{CO}$, indicating that Nb-doping decouples TRS breaking from charge order in the bulk but synchronises them near the surface. Additionally, Nb-doping raises the superconducting critical temperature $T_\mathrm{C}$ from 2.5 K to 4.4 K. Applying hydrostatic pressure enhances both $T_\mathrm{C}$ and the superfluid density by a factor of two, with a critical pressure $p_\mathrm{cr}$ ${\simeq}$ 0.85 GPa, suggesting competition with charge order. Notably, above $p_\mathrm{cr}$, we observe nodeless electron pairing and weak internal fields below $T_\mathrm{C}$, indicating broken TRS in the superconducting state. Overall, these results demonstrate a highly unconventional normal state with a depth-tunable onset of TRS breaking at ambient pressure, a transition to TRS-breaking superconductivity under low hydrostatic pressure, and an unconventional scaling between $T_\mathrm{C}$ and the superfluid density.

In this paper, we apply the framework of optimal transport to the formulation of optimal design problems. By considering the Wasserstein space as a set of design variables, we associate each probability measure with a shape configuration of a material in some ways. In particular, we focus on connections between differentials on the Wasserstein space and sensitivities in the standard setting of shape and topology optimization in order to regard the optimization procedure of those problems as gradient flows on the Wasserstein space., Comment: 15 pages

We present observations of the gamma-ray binary PSR B1259-63/LS 2883 with the Atacama Large Millimeter/submillimeter Array (ALMA) at Bands 3 (97 GHz), 6 (233 GHz), and 7 (343 GHz). PSR B1259-63/LS 2883 consists of a pulsar in a highly eccentric orbit around a massive companion star, with the pulsar passing through the circumstellar disk near periastron. Our new data were obtained over several epochs, ranging from -61 to +29 days from the periastron passage in 2024. We report an increase in flux in all bands near the periastron. The significant change in Band 3 flux suggests synchrotron emission from the interaction between the pulsar wind and the stellar wind or disk. The Band 6 flux shows an increase around periastron and a transition from thermal emission from the circumstellar disk to synchrotron emission. The Band 7 observation +24 days after periastron shows a brightening, suggesting that the pulsar's passage through the disk does not result in its immediate destruction. We discuss the implications of these results for the interaction between the pulsar wind and the circumstellar disk, such as the possible disk expansion after periastron., Comment: Accepted for publication in ApJ Letters

To develop high-performing Visual Language Models (VLMs), it is essential to prepare multimodal resources, such as image-text pairs, interleaved data, and instruction data. While multimodal resources for English are abundant, there is a significant lack of corresponding resources for non-English languages, such as Japanese. To address this problem, we take Japanese as a non-English language and propose a method for rapidly creating Japanese multimodal datasets from scratch. We collect Japanese image-text pairs and interleaved data from web archives and generate Japanese instruction data directly from images using an existing VLM. Our experimental results show that a VLM trained on these native datasets outperforms those relying on machine-translated content., Comment: 15 pages, 7 figures

Materials with axis-dependent conduction polarity are known as $p\times n$-type or goniopolar conductors that can be used for transverse thermoelectric devices, allowing the longitudinal thermal current to be converted into the transverse electrical current. Here, we performed experimental and computational studies on the transport properties of WSi$_2$ single crystals, in which such axis-dependent conduction polarity of the thermopower and the Hall coefficient have recently been reported, and demonstrated the transverse thermoelectric effect by applying the temperature gradient in a 45$^{\circ}$ rotated direction from the crystallographic axis. We have observed strongly sample-dependent transport properties, which have also been observed in previous studies, and together with first-principles calculations we show that such sample-dependent transport properties originate from the band-dependent scattering rates of carriers. The calculated band-resolved Peltier conductivity shows that the mixed-dimensional electronic structure consisting of a quasi-one-dimensional electron Fermi surface and a quasi-two-dimensional hole surface is a key property for the axis-dependent conduction polarity. The directly obtained transverse thermoelectric figure of merit is comparable to that of the anomalous Nernst materials, implying that the present material is a potential candidate for transverse thermoelectric conversion., Comment: 10 pages, 8 figures. Accepted for publication in PRX Energy

Tokenization is the first step in modern neural language model pipelines where an input text is converted to a sequence of subword tokens. We introduce from first principles a finite-state transduction framework which can efficiently encode all possible tokenizations of a regular language. We then constructively show that Byte-Pair Encoding (BPE) and MaxMatch (WordPiece), two popular tokenization schemes, fit within this framework. For BPE, this is particularly surprising given its resemblance to context-free grammar and the fact that it does not tokenize strings from left to right. An application of this is to guided generation, where the outputs of a language model are constrained to match some pattern. Here, patterns are encoded at the character level, which creates a mismatch between the constraints and the model's subword vocabulary. While past work has focused only on constraining outputs without regard to the underlying tokenization algorithm, our framework allows for simultaneously constraining the model outputs to match a specified pattern while also adhering to the underlying tokenizer's canonical tokenization., Comment: 10 pages + 5 pages in appendix

Acceleration of positive muons from thermal energy to $100~$keV has been demonstrated. Thermal muons were generated by resonant multi-photon ionization of muonium atoms emitted from a sheet of laser-ablated aerogel. The thermal muons were first electrostatically accelerated to $5.7~$keV, followed by further acceleration to 100 keV using a radio-frequency quadrupole. The transverse normalized emittance of the accelerated muons in the horizontal and vertical planes were $0.85 \pm 0.25 ~\rm{(stat.)}~^{+0.22}_{-0.13} ~\rm{(syst.)}~\pi~$mm$\cdot$mrad and $0.32\pm 0.03~\rm{(stat.)} ^{+0.05}_{-0.02} ~\rm{(syst.)}~\pi~$mm$\cdot$mrad, respectively. The measured emittance values demonstrated phase space reduction by a factor of $2.0\times 10^2$ (horizontal) and $4.1\times 10^2$ (vertical) allowing good acceleration efficiency. These results pave the way to realize the first-ever muon accelerator for a variety of applications in particle physics, material science, and other fields.

The isotopic compositions of samples returned from Cb-type asteroid Ryugu and Ivuna-type (CI) chondrites are distinct from other carbonaceous chondrites, which has led to the suggestion that Ryugu and CI chondrites formed in a different region of the accretion disk, possibly around the orbits of Uranus and Neptune. We show that, like for Fe, Ryugu and CI chondrites also have indistinguishable Ni isotope anomalies, which differ from those of other carbonaceous chondrites. We propose that this unique Fe and Ni isotopic composition reflects different accretion efficiencies of small FeNi metal grains among the carbonaceous chondrite parent bodies. The CI chondrites incorporated these grains more efficiently, possibly because they formed at the end of the disk's lifetime, when planetesimal formation was also triggered by photoevaporation of the disk. Isotopic variations among carbonaceous chondrites may thus reflect fractionation of distinct dust components from a common reservoir, implying CI chondrites and Ryugu may have formed in the same region of the accretion disk as other carbonaceous chondrites., Comment: Published open access in Science Advances

X-ray-induced carrier dynamics in silicon and gallium arsenide were investigated through intensity variations of transmitted terahertz (THz) pulses in the pico to microsecond time scale with X-ray free-electron laser and synchrotron radiation. We observed a steep reduction in THz transmission with a picosecond scale due to the X-ray-induced carrier generation, followed by a recovery on a nano to microsecond scale caused by the recombination of carriers. The rapid response in the former process is applicable to a direct determination of temporal overlap between THz and X-ray pulses for THz pump-X-ray probe experiments with an accuracy of a few picoseconds.

A search for proton decay into $e^+/\mu^+$ and a $\eta$ meson has been performed using data from a 0.373 Mton$\cdot$year exposure (6050.3 live days) of Super-Kamiokande. Compared to previous searches this work introduces an improved model of the intranuclear $\eta$ interaction cross section, resulting in a factor of two reduction in uncertainties from this source and $\sim$10\% increase in signal efficiency. No significant data excess was found above the expected number of atmospheric neutrino background events resulting in no indication of proton decay into either mode. Lower limits on the proton partial lifetime of $1.4\times\mathrm{10^{34}~years}$ for $p\rightarrow e^+\eta$ and $7.3\times\mathrm{10^{33}~years}$ for $p\rightarrow \mu^+\eta$ at the 90$\%$ C.L. were set. These limits are around 1.5 times longer than our previous study and are the most stringent to date.

Scientific machine learning (SciML) is an interdisciplinary research field that integrates machine learning, particularly deep learning, with physics theory to understand and predict complex natural phenomena. By incorporating physical knowledge, SciML reduces the dependency on observational data, which is often limited in the natural sciences. In this article, the fundamental concepts of SciML, its applications in seismology, and prospects are described. Specifically, two popular methods are mainly discussed: physics-informed neural networks (PINNs) and neural operators (NOs). PINNs can address both forward and inverse problems by incorporating governing laws into the loss functions. The use of PINNs is expanding into areas such as simultaneous solutions of differential equations, inference in underdetermined systems, and regularization based on physics. These research directions would broaden the scope of deep learning in natural sciences. NOs are models designed for operator learning, which deals with relationships between infinite-dimensional spaces. NOs show promise in modeling the time evolution of complex systems based on observational or simulation data. Since large amounts of data are often required, combining NOs with physics-informed learning holds significant potential. Finally, SciML is considered from a broader perspective beyond deep learning: statistical (or mathematical) frameworks that integrate observational data with physical principles to model natural phenomena. In seismology, mathematically rigorous Bayesian statistics has been developed over the past decades, whereas more flexible and scalable deep learning has only emerged recently. Both approaches can be considered as part of SciML in a broad sense. Theoretical and practical insights in both directions would advance SciML methodologies and thereby deepen our understanding of earthquake phenomena., Comment: English translation of the manuscript submitted to Zisin (Journal of the Seismological Society of Japan)

In this article, we investigate sequences of discontinuous martingales on submanifolds of higher-dimensional Euclidean space. Those sequences naturally arise when we deal with a sequence of harmonic maps with respect to non-local Dirichlet forms, such as fractional harmonic maps. We prove that the semimartingale topology is equivalent to the topology of locally uniform convergence in probability on the space of discontinuous martingales on manifolds. In particular, we show that the limit of any sequence of discontinuous martingales on a compact Riemannian manifold, with respect to the topology of locally uniform convergence in probability, is a martingale on the manifold., Comment: 20 pages

We study the half-indices of $\mathcal{N}=4$ super Yang-Mills theories of orthogonal and symplectic gauge groups. We find precise matching pairs of the half-indices as strong evidence of dualities of the half-BPS boundary conditions and interfaces. The half-indices encode the spectra of excitations on the holographically dual orbifold ETW brane and giant gravitons. The exact form of the index for the orbifold ETW giant gravitons is obtained from the giant graviton expansion of the half-index., Comment: 53 pages, 8 figures, v2: published version in JHEP

We examine the giant graviton expansions of the Coulomb and Higgs indices for the M2-brane SCFTs to find the closed-form expressions for the indices that encode the spectra of the $1/4$-BPS M5-brane giant gravitons and the $1/3$-BPS orbifold M5-brane giant gravitons. Consequently, we get exact forms of the twisted indices for the 6d $(2,0)$ theories describing a stack of $N$ M5-branes which generalize the unrefined indices. We confirm that they are also beautifully expanded with respect to the indices for the M2-brane giant gravitons which are obtained from the Coulomb and Higgs indices for the M2-brane SCFTs upon the change of variables., Comment: 54 pages, v2: published version in JHEP

We theoretically study nonlinear processes driven by a DC spin-electric field in the antiferromagnetic spin-1 Heisenberg model starting from the ground state in the Haldane phase. The DC spin-electric field generates finite spin current and accumulation since the symmetry protected topological order is destroyed by the field. We find two microscopic mechanisms responsible for the breakdown and a crossover between them: In weak fields, tripron-antitripron pair creation occurs through the tunneling mechanism, and in strong fields, the system is described by an effective Hamiltonian breaking the protecting symmetries. We analyze the numerically obtained results in terms of the Dykhne--Davis--Pechukas theory and Floquet theory, verifying the universal picture of the many-body Keldysh crossover in the DC-driven quantum spin systems., Comment: 8pages, 5figures

Subword regularization, used widely in NLP, improves model performance by reducing the dependency on exact tokenizations, augmenting the training corpus, and exposing the model to more unique contexts during training. BPE and MaxMatch, two popular subword tokenization schemes, have stochastic dropout regularization variants. However, there has not been an analysis of the distributions formed by them. We show that these stochastic variants are heavily biased towards a small set of tokenizations per word. If the benefits of subword regularization are as mentioned, we hypothesize that biasedness artificially limits the effectiveness of these schemes. Thus, we propose an algorithm to uniformly sample tokenizations that we use as a drop-in replacement for the stochastic aspects of existing tokenizers, and find that it improves machine translation quality., Comment: 4 pages + 4 page appendix. 3 figures

Mixture of Experts (MoE) offers remarkable performance and computational efficiency by selectively activating subsets of model parameters. Traditionally, MoE models use homogeneous experts, each with identical capacity. However, varying complexity in input data necessitates experts with diverse capabilities, while homogeneous MoE hinders effective expert specialization and efficient parameter utilization. In this study, we propose a novel Heterogeneous Mixture of Experts (HMoE), where experts differ in size and thus possess diverse capacities. This heterogeneity allows for more specialized experts to handle varying token complexities more effectively. To address the imbalance in expert activation, we propose a novel training objective that encourages the frequent activation of smaller experts, enhancing computational efficiency and parameter utilization. Extensive experiments demonstrate that HMoE achieves lower loss with fewer activated parameters and outperforms conventional homogeneous MoE models on various pre-training evaluation benchmarks. Codes will be released upon acceptance.

We study the line defect half-indices of 3d $\mathcal{N}=2$ supersymmetric Chern-Simons (CS) theories with (special)unitary, symplectic, orthogonal and exceptional gauge groups. We find that they have several beautiful infinite product $q$-series expressions in terms of Ramanujan's general theta function. For the theories with fundamental chiral multiplets, the pairs of the Neumann half-indices and one-point functions of the fundamental Wilson lines form a basis for the line defect indices in terms of the Rogers-Ramanujan type functions. Furthermore, the theories with an adjoint chiral admit the expressions as the eta-products. In particular, for the $SU(N)_{-2N}$ CS theory, there is a one-to-one correspondence between the BPS boundary local operators and the $N$-core partitions., Comment: 68 pages

Identifying reaction coordinates (RCs) is a key to understanding the mechanism of reactions in complex systems. Deep neural network (DNN) and machine learning approaches have become a powerful tool to find the RC. On the other hand, the hyperparameters that determine the DNN model structure can be highly flexible and are often selected intuitively and in a non-trivial and tedious manner. Furthermore, how the hyperparameter choice affects the RC quality remains obscure. Here, we explore the hyperparameter space by developing the hyperparameter tuning approach for the DNN model for RC and investigate how the parameter set affects the RC quality. The DNN model is built to predict the committor along the RC from various collective variables by minimizing the cross-entropy function; the hyperparameters are automatically determined using the Bayesian optimization method. The approach is applied to study the isomerization of alanine dipeptide in vacuum and in water, and the features that characterize the RC are extracted using the explainable AI (XAI) tools. The results show that the DNN models with diverse structures can describe the RC with similar accuracy, and furthermore, the features analyzed by XAI are highly similar. This indicates that the hyperparameter space is multimodal. The electrostatic potential from the solvent to the hydrogen H18 plays an important role in the RC in water. The current study shows that the structure of the DNN models can be rather flexible, while the suitably optimized models share the same features; therefore, a common mechanism from the RC can be extracted., Comment: 11 pages, 11 figures, 2 table for main text, 12 pages for supplementary material

A prototypical charge-transfer type Mott insulator NiS2 pyrite exhibits a metal-insulator transition with bandwidth control. Recent discoveries on surface-specific electronic states on other 3d transition-metal disulfide pyrites motivate us to further investigate the surface of NiS2, where metallic surface conduction is discussed. Here, the spectroscopic-imaging scanning-tunneling-microscopy observations revealed that the surface is not metallic, contrary to the expectation. Instead, the Mott gap is closed near step edges, suggesting possible electrical conduction from one-dimensional channels. The edge anomaly was observed irrespective of its magnetic order and is limited to the insulator phases., Comment: 5 pages, 5 figures

Artificial intelligence (AI)-based weather prediction research is growing rapidly and has shown to be competitive with the advanced dynamic numerical weather prediction models. However, research combining AI-based weather prediction models with data assimilation remains limited partially because long-term sequential data assimilation cycles are required to evaluate data assimilation systems. This study proposes using ensemble data assimilation for diagnosing AI-based weather prediction models, and marked the first successful implementation of ensemble Kalman filter with AI-based weather prediction models. Our experiments with an AI-based model ClimaX demonstrated that the ensemble data assimilation cycled stably for the AI-based weather prediction model using covariance inflation and localization techniques within the ensemble Kalman filter. While ClimaX showed some limitations in capturing flow-dependent error covariance compared to dynamical models, the AI-based ensemble forecasts provided reasonable and beneficial error covariance in sparsely observed regions. In addition, ensemble data assimilation revealed that error growth based on ensemble ClimaX predictions was weaker than that of dynamical NWP models, leading to higher inflation factors. A series of experiments demonstrated that ensemble data assimilation can be used to diagnose properties of AI weather prediction models such as physical consistency and accurate error growth representation.

Kagome superconductor CsV3Sb5, which exhibits intertwined unconventional charge density wave (CDW) and superconductivity, has garnered significant attention recently. Despite extensive static studies, the nature of these exotic electronic orders remains elusive. In this study, we investigate the non-equilibrium electronic structure of CsV3Sb5 via time- and angle-resolved photoemission spectroscopy. Our results reveal that upon laser excitation, the van Hove singularities immediately shift towards the Fermi level and subsequently oscillate in sync with a 1.3 THz coherent phonon mode. By analyzing the coherent intensity oscillations in the energy-momentum (E-k) map, we find that this coherent phonon is strongly coupled with electronic bands from both Sb and V orbitals. While typically observable only in the CDW state, remarkably, we find that the 1.3-THz coherent phonon mode can be persistently excited at temperatures above T_CDW, suggesting the potential existence of fluctuated CDW in CsV3Sb5. These findings enhance our understanding of the unconventional CDW control of kagome superconductivity., Comment: 11 pages, 5 figures

Recent advancements in circuit quantum electrodynamics have enabled precise manipulation and detection of the single energy quantum in quantum systems. A quantum circuit refrigerator (QCR) is capable of electrically cooling the excited population of quantum systems, such as superconducting resonators and qubits, through photon-assisted tunneling of quasi-particles within a superconductor-insulator-normal metal junction. In this study, we demonstrated instantaneous QCR in the quantum regime. We performed the time-resolved measurement of the QCR-induced cooling of photon number inside the superconducting resonator by harnessing a qubit as a photon detector. From the enhanced photon loss rate of the resonator estimated from the amount of the AC Stark shift, the QCR was shown to have a cooling power of approximately 300 aW. Furthermore, even below the single energy quantum, the QCR can reduce the number of photons inside the resonator with 100 ns pulse from thermal equilibrium. Numerical calculations based on the Lindblad master equation successfully reproduced these experimental results., Comment: 15 pages, 9 figures, and 1 table

We study the effect of irradiation from two accretion disks (minidisks) around respective black holes of stellar to intermediate masses in a circular binary on the spectrum of a circumbinary disk (CBD) surrounding them. We assume the CBD to be a standard disk and adopt the orbit-averaged irradiation flux because the viscous timescale is much longer than the orbital period. We then solve the energy equation both analytically and numerically to compute the CBD temperature distribution and the corresponding disk spectrum. We find that the analytically calculated spectra are in good agreement with the numerical ones. The CBD spectrum is almost independent of the binary mass ratio. We also find that the combined spectra of two minidisks and the CBD have double peaks, one peak in the soft X-ray band and the other in the infrared (IR) band. The former peak comes from the two minidisks, while the latter peak from the CBD. The observed flux density increases with frequency as $\nu^{1/3}$ towards the soft X-ray peak, while it decreases with frequency away from the IR peak as $\nu^{-5/3}$. The latter feature is testable with near-IR observations with Subaru and JWST., Comment: 19 pages, 7 figures. Submitted to ApJ

This paper introduces LLM-jp, a cross-organizational project for the research and development of Japanese large language models (LLMs). LLM-jp aims to develop open-source and strong Japanese LLMs, and as of this writing, more than 1,500 participants from academia and industry are working together for this purpose. This paper presents the background of the establishment of LLM-jp, summaries of its activities, and technical reports on the LLMs developed by LLM-jp. For the latest activities, visit https://llm-jp.nii.ac.jp/en/.

Warning: This paper contains examples of stereotypes and biases. Large Language Models (LLMs) exhibit considerable social biases, and various studies have tried to evaluate and mitigate these biases accurately. Previous studies use downstream tasks as prompts to examine the degree of social biases for evaluation and mitigation. While LLMs' output highly depends on prompts, previous studies evaluating and mitigating bias have often relied on a limited variety of prompts. In this paper, we investigate the sensitivity of LLMs when changing prompt variations (task instruction and prompt, few-shot examples, debias-prompt) by analyzing task performance and social bias of LLMs. Our experimental results reveal that LLMs are highly sensitive to prompts to the extent that the ranking of LLMs fluctuates when comparing models for task performance and social bias. Additionally, we show that LLMs have tradeoffs between performance and social bias caused by the prompts. Less bias from prompt setting may result in reduced performance. Moreover, the ambiguity of instances is one of the reasons for this sensitivity to prompts in advanced LLMs, leading to various outputs. We recommend using diverse prompts, as in this study, to compare the effects of prompts on social bias in LLMs.