Search

Researchers and practitioners in natural language processing and computational linguistics frequently observe and analyze the real language usage in large-scale corpora. For that purpose, they often employ off-the-shelf pattern-matching tools, such as grep, and keyword-in-context concordancers, which is widely used in corpus linguistics for gathering examples. Nonetheless, these existing techniques rely on surface-level string matching, and thus they suffer from the major limitation of not being able to handle orthographic variations and paraphrasing -- notable and common phenomena in any natural language. In addition, existing continuous approaches such as dense vector search tend to be overly coarse, often retrieving texts that are unrelated but share similar topics. Given these challenges, we propose a novel algorithm that achieves \emph{soft} (or semantic) yet efficient pattern matching by relaxing a surface-level matching with word embeddings. Our algorithm is highly scalable with respect to the size of the corpus text utilizing inverted indexes. We have prepared an efficient implementation, and we provide an accessible web tool. Our experiments demonstrate that the proposed method (i) can execute searches on billion-scale corpora in less than a second, which is comparable in speed to surface-level string matching and dense vector search; (ii) can extract harmful instances that semantically match queries from a large set of English and Japanese Wikipedia articles; and (iii) can be effectively applied to corpus-linguistic analyses of Latin, a language with highly diverse inflections., Comment: Accepted at ICLR2025

High-speed optical receivers are crucial in modern optical communication systems. While complex photonic integrated circuits (PICs) are widely employed to harness the full degrees of freedom (DOFs) of light for efficient data transmission, their waveguide nature inherently constrains two-dimensional spatial scaling to accommodate a large number of optical signals in parallel. Here, we present a novel optical receiver platform that fully exploits the high spatial parallelism and ultrabroad bandwidth of light, while leveraging all DOFs - intensity, phase, and polarization. Our solution integrates a thin metasurface, composed of silicon nanoposts, with ultrafast membrane photodetectors on a compact chip. The metasurface provides all the functionalities of conventional PICs for normal-incident spatially parallelized light, enabling high-speed detection of optical signals in various modulation formats, including simultaneous detection of 320-gigabits-per-second four-channel signals and coherent detection of a 240-gigabits-per-second signal., Comment: 18 pages, 5+6 figures

We propose BGM2Pose, a non-invasive 3D human pose estimation method using arbitrary music (e.g., background music) as active sensing signals. Unlike existing approaches that significantly limit practicality by employing intrusive chirp signals within the audible range, our method utilizes natural music that causes minimal discomfort to humans. Estimating human poses from standard music presents significant challenges. In contrast to sound sources specifically designed for measurement, regular music varies in both volume and pitch. These dynamic changes in signals caused by music are inevitably mixed with alterations in the sound field resulting from human motion, making it hard to extract reliable cues for pose estimation. To address these challenges, BGM2Pose introduces a Contrastive Pose Extraction Module that employs contrastive learning and hard negative sampling to eliminate musical components from the recorded data, isolating the pose information. Additionally, we propose a Frequency-wise Attention Module that enables the model to focus on subtle acoustic variations attributable to human movement by dynamically computing attention across frequency bands. Experiments suggest that our method outperforms the existing methods, demonstrating substantial potential for real-world applications. Our datasets and code will be made publicly available.

Diffusion in a heterogeneous environment or diffusion of a particle that shows conformational fluctuations can be described by Brownian motions with stochastic diffusion coefficients (DCs). In the present study, we investigate first passage time (FPT) properties of diffusion with a broad class of stochastic DCs that are positive and non-zero. We show that for diffusion in one-dimensional semi-infinite domain with an absorbing boundary, the eventual absorption is certain. We also show that compared to diffusion whose DC is the ensemble average of a stochastic DC, in diffusion with the stochastic DC, there are particles that reach the absorbing boundary earlier. The approximate proportion of preceding particles is determined by the probability distribution of the time average of a stochastic DC for a given distance to the absorbing boundary. In addition, when particles begin to reach the absorbing boundary before the time change in a stochastic DC occurs, a stochastic DC with a larger supremum leads to an earlier arrival of particles at the absorbing boundary even if the ensemble averages of stochastic DCs are the same. For ergodic DCs, three more properties are revealed. The mean FPT is infinite. In addition, if particles take a long time to reach the absorbing boundary, the exact proportion of preceding particles is almost zero and the FPT distribution can be approximated by the L\'evy-Smirnov distribution. We show that these three properties of diffusion with an ergodic DC result from the convergence of the time average of the DC to the ensemble average. For three-dimensional diffusion with a spherical absorbing boundary, we obtain essentially the same results, except that the eventual absorption is not certain. Our results suggest that fluctuations in a DC play an important role in diffusion-limited reactions triggered by single molecules in physics, chemistry, or biology., Comment: 10 pages, 4 figures

A neutron whispering gallery state is a quantum state localized on a material surface bound by the centrifugal force and the material potential. Precise measurements of such quantum states enable tests of quantum mechanics in non-inertial frames, characterization of the surface potential, and searches for hypothetical short-range interactions at the nanometer scale. We observed a neutron whispering gallery state on a $\mathrm{SiO}_2$ concave mirror using a pulsed cold neutron beam. The measured results agree with theoretical calculations within $1.9\%$ for the centrifugal acceleration $a \approx 7\times 10^7\,\mathrm{m/s^2}$, which is due to unmodeled deviations of the shape of the concave mirror edge from an ideal one. We found that the sensitivity itself was $1\times 10^{-4}$, which is two orders of magnitude better than the above agreement., Comment: 9 pages, 6 figures, submitted to PRD; typos corrected

Vertex operator algebras (VOAs) arise in protected subsectors of supersymmetric quantum field theories, notably in 4d ${\mathcal N}=2$ superconformal field theories (SCFT) via the Schur sector and in twisted 3d ${\mathcal N}=4$ theories via boundary algebras. These constructions are connected through twisted circle compactifications, which can be best understood from the dynamics of BPS particles in the Coulomb branch of the 4d SCFT. This data is encoded in an operator $\hat\Phi$ acting on the Hilbert space of an auxiliary quantum mechanics of BPS particles, whose trace yields the partition functions of a 3d topological field theory (TFT) bounding the VOA. We generalize this trace formula by considering higher powers of $\hat\Phi$, leading to a finite family of VOAs associated with a given 4d SCFT. Applying this framework to Argyres-Douglas theories labeled by $(A_1, G)$, where $G$ is an ADE-type group of rank up to 8, we extract the modular data of the family of boundary VOAs via TFT partition function calculations on Seifert manifolds. Our results suggest that the modular data obtained from different powers of $\hat\Phi$ are related by Galois transformations., Comment: 69 pages

PET is affected by statistical noise due to constraints on tracer dose and scan duration, impacting both diagnostic performance and quantitative accuracy. While deep learning (DL)-based PET denoising methods have been used to improve image quality, they may introduce over-smoothing, compromising quantitative accuracy. We propose a method for making a DL solution more reliable and apply it to the conditional deep image prior (DIP). We introduce the idea of stability information in the optimization process of conditional DIP, enabling the identification of unstable regions within the network's optimization trajectory. Our method incorporates a stability map, which is derived from multiple intermediate outputs of moderate network at different optimization steps. The final denoised image is then obtained by computing linear combination of the DIP output and the original reconstructed image, weighted by the stability map. Our method effectively reduces noise while preserving small structure details in brain FDG images. Results demonstrated that our approach outperformed existing methods in peak-to-valley ratio and noise suppression across various low-dose levels. Region-of-interest analysis confirmed that the proposed method maintains quantitative accuracy without introducing under- or over-estimation. We applied our method to full-dose PET data to assess its impact on image quality. The results revealed that the proposed method significantly reduced background noise while preserving the peak-to-valley ratio at a level comparable to that of unfiltered full-dose PET images. The proposed method introduces a robust approach to DL-based PET denoising, enhancing its reliability and preserving quantitative accuracy. This strategy has the potential to advance performance in high-sensitivity PET scanners, demonstrating that DL can extend PET imaging capabilities beyond low-dose applications., Comment: 10 pages, 8 figures

Spin-Orbitronics leverages the spin and orbital degrees of freedom in solids for information processing. The orbital Edelstein effect and orbital Hall effect, where the charge current induces a nonequilibrium orbital angular momentum, offer a promising method to manipulate nanomagnets efficiently using light elements. Despite extensive research, understanding the Onsager reciprocity of orbital transport, fundamentally rooted in the second law of thermodynamics and time-reversal symmetry, remains elusive. In this study, we experimentally demonstrate the Onsager reciprocity of orbital transport in an orbital Edelstein system by utilizing nonlocal measurements. This method enables the precise identification of the chemical potential generated by orbital accumulation, avoiding the limitations associated with local measurements. Remarkably, we observe that the direct and inverse orbital-charge conversion processes produce identical electric voltages, confirming Onsager reciprocity in orbital transport. Additionally, we find that the orbital decay length, approximately 100 nm at room temperature, is independent of Cu thickness and decreases with lowering temperature, revealing a distinct contrast to spin transport behavior. Our findings provide valuable insights into both the reciprocity of the charge-orbital interconversion and the nonlocal correlation of orbital degree of freedom, laying the ground for orbitronics devices with long-range interconnections.

Evaluating writing quality is complex and time-consuming often delaying feedback to learners. While automated writing evaluation tools are effective for English, Korean automated writing evaluation tools face challenges due to their inability to address multi-view analysis, error propagation, and evaluation explainability. To overcome these challenges, we introduce UKTA (Unified Korean Text Analyzer), a comprehensive Korea text analysis and writing evaluation system. UKTA provides accurate low-level morpheme analysis, key lexical features for mid-level explainability, and transparent high-level rubric-based writing scores. Our approach enhances accuracy and quadratic weighted kappa over existing baseline, positioning UKTA as a leading multi-perspective tool for Korean text analysis and writing evaluation., Comment: Accepted by SAC 2025

The Soft X-ray Imager (SXI) is the X-ray charge-coupled device (CCD) camera for the soft X-ray imaging telescope Xtend installed on the X-ray Imaging and Spectroscopy Mission (XRISM), which was adopted as a recovery mission for the Hitomi X-ray satellite and was successfully launched on 2023 September 7 (JST). In order to maximize the science output of XRISM, we set the requirements for Xtend and find that the CCD set employed in the Hitomi/SXI or similar, i.e., a $2 \times 2$ array of back-illuminated CCDs with a $200~\mu$m-thick depletion layer, would be practically best among available choices, when used in combination with the X-ray mirror assembly. We design the XRISM/SXI, based on the Hitomi/SXI, to have a wide field of view of $38' \times 38'$ in the $0.4-13$ keV energy range. We incorporated several significant improvements from the Hitomi/SXI into the CCD chip design to enhance the optical-light blocking capability and to increase the cosmic-ray tolerance, reducing the degradation of charge-transfer efficiency in orbit. By the time of the launch of XRISM, the imaging and spectroscopic capabilities of the SXI has been extensively studied in on-ground experiments with the full flight-model configuration or equivalent setups and confirmed to meet the requirements. The optical blocking capability, the cooling and temperature control performance, and the transmissivity and quantum efficiency to incident X-rays of the CCDs are also all confirmed to meet the requirements. Thus, we successfully complete the pre-flight development of the SXI for XRISM., Comment: 14 pages, 11 figures, 3 tables, Accepted for publication in PASJ XRISM special issue

Gauge origami is a generalized supersymmetric gauge theory defined on several intersecting space-time components. It provides a systematic way to consider generalizations of instantons. In this thesis, we explore the gauge origami system in $\mathbb{R}^{1,1}\times \mathbb{C}^{4}$ and its BPS/CFT correspondence. String theoretically, instantons of the gauge origami system arise from D0-branes bound to D$(2p)$-branes wrapping cycles in $\mathbb{C}^{4}$. The low energy theory of the D0-branes is understood as an $\mathcal{N}=2$ supersymmetric quiver quantum mechanical system and the Witten index of it produces the instanton partition function. We define a $q$-deformed quiver Cartan matrix associated to this quiver structure and introduce vertex operators associated with the D-branes and show that the contour integral formula for the Witten index has a nice free field realization. Such free field realization leads to the concept of BPS $qq$-characters or BPS quiver W-algebras, which are generalizations of the conventional deformed W-algebras. The $qq$-characters of D2 and D4-branes correspond to screening charges and generators of the affine quiver W-algebra, respectively. On the other hand, the $qq$-characters of D6 and D8-branes represent novel types of $qq$-characters, where monomial terms are characterized by plane partitions and solid partitions. The composition of these $qq$-characters yields the instanton partition functions of the gauge origami system, eventually establishing the BPS/CFT correspondence. Additionally, we demonstrate that the fusion of $qq$-characters of D-branes in lower dimensions results in higher-dimensional D-brane $qq$-characters. We also investigate quadratic relations among these $qq$-characters. Furthermore, we explore the relationship with the representations of the quantum toroidal $\mathfrak{gl}_{1}$., Comment: PhD thesis submitted to Department of Physics, University of Tokyo (Dec. 2024), 174 pages. Based on arXiv:2310.08545 and arXiv:2411.01987

Mixture-of-Experts (MoE) model architecture has emerged as a promising solution for scaling transformer models efficiently, offering sparse activation that reduces computational costs while increasing model capacity. However, as MoE models scale, they need to be distributed across GPU devices, thus face critical performance bottlenecks due to their large memory footprint. Expert parallelism distributes experts across GPUs, however, faces key challenges including an unbalanced token routing and expert activation, resulting in communication tail latency and processing inefficiencies. While existing solutions address some of these issues, they fail to resolve the dual challenges of load imbalance and communication skew. The imbalance in token processing load across experts causes uneven processing times on different GPUs, while communication skew between GPUs leads to unbalanced inter-GPU data transfers. These factors degrade the performance of MoE models by increasing tail latency and reducing overall throughput. To address these limitations, we propose an Integer Linear Programming (ILP) formulation to optimize expert placement by jointly considering token load, communication, and computation costs. We exploit the property that there is a token routing dependency across layers, where tokens routed to a specific expert in one layer are likely to be routed to a limited set of experts in the subsequent layer. Our solution, MoETuner, offers an optimal expert-to-GPU assignment that minimizes inter-GPU token routing costs and balances token processing across devices, thereby reducing tail latency and end-to-end execution time. Experimental results demonstrate 9.3% and 17.5% of end-to-end speedups for single-node and multi-node inference respectively, showcasing the potential of our ILP-based optimization for offering expert parallel solutions for next-generation MoEs.

We apply a voltage pulse to electrically excite the incompressible region of a two-dimensional electron liquid in the $\nu=2/3$ fractional quantum Hall state and investigate the collective excitations in both the edge and bulk via photoluminescence spectral energy shifts. Introducing an offset in the voltage pulse significantly enhances the excitation signal. Real-space and time-resolved measurements reveal the dynamics of the bulk excitations, with an estimated group velocity of approximately $3 \times 10^4$ m/s. These bulk excitations align well with the magneto-plasmon model. Our results highlight the topological link between edge and bulk states, providing a novel approach to exploring solid-state analogs of quantum gravity.

In positron emission tomography (PET), time-of-flight (TOF) information localizes source positions along lines of response. Cherenkov-radiator-integrated microchannel-plate photomultiplier tubes have achieved 30 ps TOF resolution, demonstrating cross-sectional imaging without reconstruction. Such ultrafast TOF detectors would free PET from conventional ring geometries. Therefore, this study aimed at investigating imaging characteristics of a dual-panel PET with ultrafast TOF detectors using Geant4 simulation. Two detector panels ($142 \times 142~mm^2$), which consisted of 5.0 mm-thick bismuth germanate pixelized crystals with a 5.75 mm pitch, were placed face-to-face at a 300 mm distance. Spatial resolution and image noise with various TOF resolutions from 30 to 90 ps were evaluated. Because degraded efficiency may cancel TOF gain in image quality, detection efficiency was also parameterized by reducing coincidence counts. A numerical phantom, which consisted of multi-rod and uniform cylindrical activities (21 MBq in total), was simulated for 600 s. Regarding image reconstruction, results of a backprojection (i.e., no reconstruction) were compared with those of the maximum likelihood expectation maximization (MLEM). The dual-panel PET required a 40 ps TOF resolution to have a similar spatial resolution to that of a non-TOF ring PET (300 mm in diameter) for the same detection efficiency. The TOF resolution should be 30 ps with half the efficiency to maintain a similar image noise to that of the 40 ps TOF with complete efficiency. MLEM provided better imaging performance than backprojection, even at 30 ps TOF. The feasibility of the proposed dual-panel PET with ultrafast TOF detectors was shown., Comment: 8 pages, 9 figures

The design of spin-orbit torque properties in two-dimensional (2D) materials presents one of the challenges of modern spintronics. In this context, 2D layers involving rare-earth ions $-$ which give rise to robust magnetism, exhibit pronounced orbital polarization of the states, and carry strong spin-orbit interaction $-$ hold particular promise. Here, we investigate ferromagnetic Janus H-phase monolayers of 4$f$-Eu rare-earth dichalcogenides EuSP, EuSSe, and EuSCl using first-principles calculations. We demonstrate that all compounds exhibit significant spin-orbit torques which originate predominantly in the colossal current-induced orbital response on the Eu $f$-electrons. Moreover, we demonstrate that the corresponding orbital torques can be used to drive strong in-plane currents of orbital angular momentum with non-trivial direction of orbital polarization. Our findings promote $f$-orbital-based 2D materials as a promising platform for in-plane orbital pumping and spin-orbit torque applications, and motivate further research on educated design of orbital properties for orbitronics with 2D materials., Comment: 13 pages, 4 figures

Out-of-distribution (OOD) detection is a task that detects OOD samples during inference to ensure the safety of deployed models. However, conventional benchmarks have reached performance saturation, making it difficult to compare recent OOD detection methods. To address this challenge, we introduce three novel OOD detection benchmarks that enable a deeper understanding of method characteristics and reflect real-world conditions. First, we present ImageNet-X, designed to evaluate performance under challenging semantic shifts. Second, we propose ImageNet-FS-X for full-spectrum OOD detection, assessing robustness to covariate shifts (feature distribution shifts). Finally, we propose Wilds-FS-X, which extends these evaluations to real-world datasets, offering a more comprehensive testbed. Our experiments reveal that recent CLIP-based OOD detection methods struggle to varying degrees across the three proposed benchmarks, and none of them consistently outperforms the others. We hope the community goes beyond specific benchmarks and includes more challenging conditions reflecting real-world scenarios. The code is https://github.com/hoshi23/OOD-X-Benchmarks.

According to the stages-of-inference hypothesis, early layers of language models map their subword-tokenized input, which does not necessarily correspond to a linguistically meaningful segmentation, to more meaningful representations that form the model's "inner vocabulary". Prior analysis of this detokenization stage has predominantly relied on probing and interventions such as path patching, which involve selecting particular inputs, choosing a subset of components that will be patched, and then observing changes in model behavior. Here, we show that several important aspects of the detokenization stage can be understood purely by analyzing model weights, without performing any model inference steps. Specifically, we introduce an analytical decomposition of first-layer attention in GPT-2. Our decomposition yields interpretable terms that quantify the relative contributions of position-related, token-related, and mixed effects. By focusing on terms in this decomposition, we discover weight-based explanations of attention bias toward close tokens and attention for detokenization., Comment: 22 pages, 14 figures, to appear in NAACL Findings 2025

Boson sampling stands out as a promising approach toward experimental demonstration of quantum computational advantage. However, the presence of physical noise in near-term experiments hinders the realization of the quantum computational advantage with boson sampling. Since physical noise in near-term boson sampling devices is inevitable, precise characterization of the boundary of noise rates where the classical intractability of boson sampling is maintained is crucial for quantum computational advantage using near-term devices. In this work, we identify the level of partial distinguishability noise that upholds the classical intractability of boson sampling. We find that boson sampling with on average $O(\log N)$ number of distinguishable photons out of $N$ input photons maintains the equivalent complexity to the ideal boson sampling case. By providing strong complexity theoretical evidence of the classical intractability of noisy boson sampling, we expect that our findings will enable one to ultimately demonstrate quantum computational advantage with noisy boson sampling experiments in the near future., Comment: 15 pages, 1 figure

Controlling/storing information carriers, such as electron charge and spin, is key for modern information society, and significant efforts have been paid made to establish novel technologies at the nanoscale. The rise of Si-based semiconductor technology and magnetism-based technology has been motivated by the aforementioned demands. However, both technologies have been individually developed, with little effort in fusing them. Hence, establishing a technology to bridge semiconductor and magnetism-based technologies that would allow realization of a novel information device is strongly awaited. In line with this research strategy, the creation of a magnetic device using semiconductors would enable fundamental innovation. Here, we show that a mother material for modern electronics, Si, gives rise to a giant room-temperature orbital Hall effect (OHE), enabling the creation of novel energy-efficient magnetic memory via efficient torque generation. The orbital torque efficiency largely exceeds that of the archetypal metallic materials used in the OHE. Our achievement overtures the conventional understanding that nonmagnetic semiconductors cannot play a pivotal role in magnetic devices and paves a new avenue for creating novel information devices through the fusion of semiconductor and magnetism-based technologies., Comment: 24 pages, 5 figures

Exploring and manipulating the orbital degrees of freedom in solids has become a fascinating research topic in modern magnetism. Here, we demonstrate that spin waves can provide a way to control electronic orbital magnetism by the mechanism of scalar spin chirality, allowing for experimental detection using techniques such as the magneto-optical Kerr effect and scanning transmission electron microscopy. By applying linear spin wave theory, we uncover that electronic magnon-driven orbital magnetization is extremely sensitive to the character of the magnonic excitations. Furthermore, we show that both the induced electronic orbital magnetism and the Nernst transport properties of the orbital angular momentum can be regulated by the strength of the Dzyaloshinskii-Moriya interaction, Kitaev interaction, as well as the direction and magnitude of the external magnetic field. We argue that magnon-mediated electronic orbital magnetism presents an emergent variable which has to be taken into account when considering the physics of coupling magnonic excitiations to phonons and light., Comment: 6 pages, 4 figures

The scalar spin chirality (SSC), defined as a triple product of spins, is essential for describing noncoplanar spin structures and understanding chiral physics in magnetic systems. Traditionally, SSC has been discussed primarily in the context of noncoplanar ground-state spin configurations at zero temperature, as collinear spin systems are generally thought to lack SSC. Consequently, whether the SSC can emerge at finite temperatures in spin systems with collinear ground states remains an open question and has yet to be fully understood. In this study, we theoretically demonstrate that thermally excited magnons can induce SSC even in collinear spin systems. By considering 2D ferromagnets on Kagome and honeycomb lattices, we demonstrate that the Dzyaloshinskii-Moriya interactions (DMI) which break the effective time-reversal symmetry in the magnon Hamiltonian can lead to finite SSC at finite temperatures. Using a simple spin model, we show both numerically and analytically that the SSC increases with the magnitude of DMI and temperature. Furthermore, calculations based on realistic material parameters reveal that the magnon-induced SSC can achieve a magnitude comparable to those observed in non-coplanar spin configurations. These findings suggest that SSC plays a significant role even in collinear spin systems, providing new insights into the chiral physics of magnetic materials., Comment: 10 pages, 5 figures

Chirality, an intrinsic preference for a specific handedness, is a fundamental characteristic observed in nature. In magnetism, magnetic chirality arises from the anti-symmetric Dzyaloshinskii-Moriya interaction in competition with the symmetric Heisenberg exchange interaction. Traditionally, the anti-symmetric interaction has been considered minor relative to the symmetric interaction. In this study, we demonstrate an observation of giant magnetic chirality during the dimensionality crossover of magnetic materials from three-dimensional to two-dimensional. The ratio between the anti-symmetric and symmetric interactions exhibits a reversal in their dominance over this crossover, overturning the traditional consideration. This observation is validated theoretically using a non-local interaction model and tight-binding calculation with distinct pairing schemes for each exchange interaction throughout the crossover. Additional experiments investigating the asphericity of orbital moments corroborate the robustness of our findings. Our findings highlight the critical role of dimensionality in shaping magnetic chirality and offer strategies for engineering chiral magnet states with unprecedented strength, desired for the design of spintronic materials., Comment: 21 pages, 4 figures

Superconducting diode effect (SDE) is theoretically examined in double quantum dot coupled to three superconducting leads, $L$, $R1$ and $R2$. Lead $L$ is commonly connected to two quantum dots (QD1, QD2) while lead $R1$ ($R2$) is connected to QD1 (QD2) only. The phase differences $\varphi_{1}$ between leads $L$ and $R1$ and $\varphi_{2}$ between leads $L$ and $R2$ are tuned independently. The critical current into lead $R1$ depends on its direction unless $\varphi_{2} = 0$, $\pi$, which is ascribable to the formation of Andreev molecule between the QDs. In the absence of electron-electron interaction $U$ in the QDs, the spectrum of the Andreev bound states forms Dirac cones in the $\varphi_{1}-\varphi_{2}$ plane if the energy levels in the QDs are tuned to the Fermi level in the leads. The SDE is enhanced to almost 30\% when $\varphi_{2}$ is set to the value at the Dirac points. In the presence of $U$, the SDE is still observed when $U$ is smaller than the superconducting energy gap in the leads. Our device should be one of the minimal models for the SDE since a similar device with a single QD does not show the SDE., Comment: 24 pages, 10 figures

A segmented-scintillator-based implantation detector was developed to study the energy distribution of beta-delayed neutrons emitted from exotic isotopes. The detector comprises a 34 $\times$ 34 YSO scintillator coupled to an 8 $\times$ 8 Position-Sensitive Photo-Multiplier Tube (PSPMT) via a tapered light guide. The detector was used at RIBF, RIKEN, for time-of-flight-based neutron spectroscopy measurement in the $^{78}$Ni region. The detector provides the position and timing resolution necessary for ion-beta correlations and ToF measurements. The detector provides a high $\sim$ 80 $\%$ beta-detection efficiency and a sub-nanosecond timing resolution. This contribution discusses the details of the design, operation, implementation, and analysis developed to obtain neutron time-of-flight spectrum and the analysis methods in the context of neutron-rich nuclei in the $^{78}$Ni region.

Density matrix exponentiation (DME) is a quantum algorithm that processes multiple copies of a program state $\sigma$ to realize the Hamiltonian evolution $e^{-i \sigma t}$. While serving as a prototypical sample-based quantum algorithm, DME is a powerful tool for various quantum information processing tasks, such as quantum principal component analysis and Hamiltonian simulation. In this work, we present a detailed sample complexity analysis of DME and sample-based Hamiltonian simulation. In particular, we prove that the sample complexity of DME is no larger than $4t^2/\varepsilon$, where $t$ is the desired evolution time and $\varepsilon$ is the desired imprecision level, as quantified by the normalized diamond distance. We also establish a fundamental lower bound on the sample complexity of sample-based Hamiltonian simulation, which matches our DME sample complexity bound up to a constant multiplicative factor, thereby proving that DME is optimal for sample-based Hamiltonian simulation. Finally, we point out that the DME sample complexity analysis in Appendix A of [Kimmel et al., npj Quantum Information 3, 13 (2017)] appears to be incomplete, highlighting the need for the results presented here, given the extensive use of DME over the past decade since its original proposal., Comment: 20 pages

In-person presentations commonly depend on projectors or screens, requiring input devices for slide transitions and laser pointing. This paper introduces a glove-based pointer device that integrates these functions, offering an alternative to conventional tools. The device leverages accelerometer and gyroscope technology to enhance precision and usability. We evaluated its performance by comparing it to the original CheerPod interface in hierarchical menu navigation tasks, involving participants aged 18 to 25. Results indicate task completion times ranging from 9 to 15 seconds with the proposed device, highlighting its efficiency and consistency. While the original CheerPod interface performed adequately, the glove-based pointer demonstrated advantages in reliability across tasks. These findings contribute to the design considerations for wearable input devices and suggest pathways for future improvements in presentation tools., Comment: 10 pages, 5 figures, 17 references, 4 appendix tables

Understanding the coherent interplay of light with the magnetization in metals has been a long-standing problem in ultrafast magnetism. While it is known that when laser light acts on a metal it can induce magnetization via the process known as the inverse Faraday effect (IFE), the most basic ingredients of this phenomenon are still largely unexplored. In particular, given a strong recent interest in orbital non-equilibrium dynamics and its role in mediating THz emission in transition metals, the exploration of distinct features in spin and orbital IFE is pertinent. Here, we present a first complete study of the spin and orbital IFE in $3d$, $4d$ and $5d$ transition metals of groups IV$-$XI from first-principles. By examining the dependence on the light polarization and frequency, we show that the laser-induced spin and orbital moments may vary significantly both in magnitude and sign. We underpin the interplay between the crystal field splitting and spin-orbit interaction as the key factor which determines the magnitude and key differences between the spin and orbital response. Additionally, we highlight the anisotropy of the effect with respect to the ferromagnetic magnetization and to the crystal structure. The provided complete map of IFE in transition metals is a key reference point in the field of optical magnetism.

In photonic quantum information processing, quantum operations using nonlinear photon-photon interactions are vital for implementing two-qubit gates and enabling faithful entanglement swapping. However, due to the weak interaction between single photons, the all-photonic realization of such quantum operations has remained out of reach so far. Herein, we demonstrate a first entanglement swapping using sum-frequency generation (SFG) between single photons in a $\chi^{(2)}$-nonlinear optical waveguide. We show that a highly efficient, stable SFG-based Bell-state analyzer and an ultralow-dark-count superconducting single-photon detector satisfy the high signal-to-noise ratio requirement for the swapping protocol.Furthermore, the system clock is enhanced by utilizing ultrafast telecom entangled photon pair sources that operate in the GHz range. Our results confirm a lower bound 0.770(76) for the swapped state's fidelity, surpassing the classical limit of 0.5 successfully. Our findings highlight the strong potential of broadband all-single-photonic nonlinear interactions for further sophistication in long-distance quantum communication and photonic quantum computation., Comment: 17 pages, 4 figures

Text-based generation and editing of 3D scenes hold significant potential for streamlining content creation through intuitive user interactions. While recent advances leverage 3D Gaussian Splatting (3DGS) for high-fidelity and real-time rendering, existing methods are often specialized and task-focused, lacking a unified framework for both generation and editing. In this paper, we introduce SplatFlow, a comprehensive framework that addresses this gap by enabling direct 3DGS generation and editing. SplatFlow comprises two main components: a multi-view rectified flow (RF) model and a Gaussian Splatting Decoder (GSDecoder). The multi-view RF model operates in latent space, generating multi-view images, depths, and camera poses simultaneously, conditioned on text prompts, thus addressing challenges like diverse scene scales and complex camera trajectories in real-world settings. Then, the GSDecoder efficiently translates these latent outputs into 3DGS representations through a feed-forward 3DGS method. Leveraging training-free inversion and inpainting techniques, SplatFlow enables seamless 3DGS editing and supports a broad range of 3D tasks-including object editing, novel view synthesis, and camera pose estimation-within a unified framework without requiring additional complex pipelines. We validate SplatFlow's capabilities on the MVImgNet and DL3DV-7K datasets, demonstrating its versatility and effectiveness in various 3D generation, editing, and inpainting-based tasks., Comment: Project Page: https://gohyojun15.github.io/SplatFlow/

In this study, we investigate the spin and orbital densities induced by magnetization dynamics in a planar bilayer heterostructure. To do this, we employed a theory of adiabatic pumping using the Keldysh formalism and Wigner expansion. We first conduct simulations on a model system to determine the parameters that control the spin and orbital pumping into an adjacent non-magnetic metal. We conclude that, in principle, the orbital pumping can be as significant as spin pumping when the spin-orbit coupling is present in the ferromagnet. We extend the study to realistic heterostructures involving heavy metals (W, Pt, Au) and light metals (Ti, Cu) by using first-principles calculations. We demonstrate that orbital pumping is favored in metals with $d$ states close to the Fermi level, such as Ti, Pt, and W, but is quenched in materials lacking such states, such as Cu and Au. Orbital injection is also favored in materials with strong spin-orbit coupling, leading to large orbital pumping in Ni/(Pt, W) bilayers.

The necessary time required to control a many-body quantum system is a critically important issue for the future development of quantum technologies. However, it is generally quite difficult to analyze directly, since the time evolution operator acting on a quantum system is in the form of time-ordered exponential. In this work, we examine the Baker-Campbell-Hausdorff (BCH) formula in detail and show that a distance between unitaries can be introduced, allowing us to obtain a lower bound on the control time. We find that, as far as we can compare, this lower bound on control time is tighter (better) than the standard quantum speed limits. This is because this distance takes into account the algebraic structure induced by Hamiltonians through the BCH formula, reflecting the curved nature of operator space. Consequently, we can avoid estimates based on shortcuts through algebraically impossible paths, in contrast to geometric methods that estimate the control time solely by looking at the target state or unitary operator., Comment: 27 pages, 2 figures. In order to make the paper as accessible as possible, we have added a bit lengthy appendices

The integration of Retrieval-Augmented Generation (RAG) with Multimodal Large Language Models (MLLMs) has revolutionized information retrieval and expanded the practical applications of AI. However, current systems struggle in accurately interpreting user intent, employing diverse retrieval strategies, and effectively filtering unintended or inappropriate responses, limiting their effectiveness. This paper introduces Contextual Understanding and Enhanced Search with MLLM (CUE-M), a novel multimodal search framework that addresses these challenges through a multi-stage pipeline comprising image context enrichment, intent refinement, contextual query generation, external API integration, and relevance-based filtering. CUE-M incorporates a robust filtering pipeline combining image-based, text-based, and multimodal classifiers, dynamically adapting to instance- and category-specific concern defined by organizational policies. Evaluations on a multimodal Q&A dataset and a public safety benchmark demonstrate that CUE-M outperforms baselines in accuracy, knowledge integration, and safety, advancing the capabilities of multimodal retrieval systems., Comment: Preprint. Under review

High-energy large-scale particle colliders generate data at extraordinary rates. Developing real-time high-throughput data compression algorithms to reduce data volume and meet the bandwidth requirement for storage has become increasingly critical. Deep learning is a promising technology that can address this challenging topic. At the newly constructed sPHENIX experiment at the Relativistic Heavy Ion Collider, a Time Projection Chamber (TPC) serves as the main tracking detector, which records three-dimensional particle trajectories in a volume of a gas-filled cylinder. In terms of occupancy, the resulting data flow can be very sparse reaching $10^{-3}$ for proton-proton collisions. Such sparsity presents a challenge to conventional learning-free lossy compression algorithms, such as SZ, ZFP, and MGARD. In contrast, emerging deep learning-based models, particularly those utilizing convolutional neural networks for compression, have outperformed these conventional methods in terms of compression ratios and reconstruction accuracy. However, research on the efficacy of these deep learning models in handling sparse datasets, like those produced in particle colliders, remains limited. Furthermore, most deep learning models do not adapt their processing speeds to data sparsity, which affects efficiency. To address this issue, we propose a novel approach for TPC data compression via key-point identification facilitated by sparse convolution. Our proposed algorithm, BCAE-VS, achieves a $75\%$ improvement in reconstruction accuracy with a $10\%$ increase in compression ratio over the previous state-of-the-art model. Additionally, BCAE-VS manages to achieve these results with a model size over two orders of magnitude smaller. Lastly, we have experimentally verified that as sparsity increases, so does the model's throughput., Comment: 37 pages, 12 figures, submitted to Journal of Computational Physics

In magnetic systems, angular momentum is carried by the spin and orbital degrees of freedom. Non-local devices can be used to study angular momentum transport. They consist of parallel heavy-metal nanowires placed on top of magnetic insulators like yttrium iron garnet (YIG), facilitating the transmission of information by magnons, generated by the accumulation of spin at the interface, created via the Spin Hall Effect (SHE) and detected via the inverse SHE (iSHE). It has been demonstrated that these processes have comparable efficiencies when the role of the detector and injector is reversed, which points to reciprocity of the processes. However, we show that by adding Ru as a source of direct and inverse orbital Hall effect (OHE), the system no longer exhibits this reciprocity. Specifically, the generation of magnons via the combination of SHE and OHE and detection via the iSHE is found to be about 35% more efficient than the inverse process for our system.

This paper explores the problem of 3D human pose estimation from only low-level acoustic signals. The existing active acoustic sensing-based approach for 3D human pose estimation implicitly assumes that the target user is positioned along a line between loudspeakers and a microphone. Because reflection and diffraction of sound by the human body cause subtle acoustic signal changes compared to sound obstruction, the existing model degrades its accuracy significantly when subjects deviate from this line, limiting its practicality in real-world scenarios. To overcome this limitation, we propose a novel method composed of a position discriminator and reverberation-resistant model. The former predicts the standing positions of subjects and applies adversarial learning to extract subject position-invariant features. The latter utilizes acoustic signals before the estimation target time as references to enhance robustness against the variations in sound arrival times due to diffraction and reflection. We construct an acoustic pose estimation dataset that covers diverse human locations and demonstrate through experiments that our proposed method outperforms existing approaches., Comment: Accepted at BMVC2024

We address the orbital distribution of stars in merging nuclear star clusters (NSCs) and the subsequent effects on supermassive black hole binary (SMBHB) evolution. We ran direct-summation $N$-body simulations with different initial conditions to do a detailed study of the resulting NSC after their progenitors had merged. Our findings reveal that prograde stars form a flattened structure, while retrograde stars have a more spherical distribution. The axial ratios of the prograde component vary based on the presence and mass ratio of the SMBHs. The fraction of prograde and retrograde stars depends on the merger orbital properties and the SMBH mass ratio. The interactions of retrograde stars with the SMBHB affect the eccentricity and separation evolution of the binary. Our analysis reveals a strong correlation between the angular momentum and eccentricity of the SMBH binary. This relationship could serve as a means to infer information about the stellar dynamics surrounding the binary. We find that prograde orbits are particularly close to the binary of SMBHs, a promising fact regarding extreme mass ratio inspiral (EMRI) production. Moreover, prograde and retrograde stars have different kinematic structures, with the prograde stars typically rotating faster than the retrograde ones. The line-of-sight velocity and velocity dispersion, as well as the velocity anisotropy of each NSC, depend on the initial merger orbital properties and SMBH mass ratios. The prograde and retrograde stars always show different behaviours. The distribution of stellar orbits and the dynamical properties of each kinematic population can potentially be used as a way to tell the properties of the parent nuclei apart, and has an important impact on expected rates of EMRIs, which will be detected by future gravitational wave observatories such as the Laser Interferometer Space Antenna (LISA). [abridged], Comment: 14 pages, 8 figures, 2 tables. Two appendices on zenodo (https://zenodo.org/records/13986686). Accepted for publication in A&A

Cooperatively Localizing robots should seek optimal control strategies to maximize precision of position estimation and ensure safety in flight. Observability-Aware Trajectory Optimization has strong potential to address this issue, but no concrete link between observability and precision has been proven yet. In this paper, we prove that improvement in positioning precision inherently follows from optimizing observability. Based on this finding, we develop an Observability-Aware Control principle to generate observability-optimal control strategies. We implement this principle in a Model Predictive Control framework, and we verify it on a team of quadrotor Unmanned Aerial Vehicles comprising a follower vehicle localizing itself by tracking a leader vehicle in both simulations and real-world flight tests. Our results demonstrate that maximizing observability contributed to improving global positioning precision for the quadrotor team., Comment: 12 pages, 5 figures

Convergence of a matrix decomposition technique, the multi-field singular value decomposition (MFSVD) which efficiently analyzes nonlinear correlations by simultaneously decomposing multiple fields, is investigated. Toward applications in turbulence studies, we demonstrate that SVD for an artificial matrix with multi-scale structures reproduces the power-law-like distribution in the singular value spectrum with several orthogonal modes. Then, MFSVD is applied to practical turbulence field data produced by numerical simulations. It is clarified that relative errors in the reproduction of quadratic nonlinear quantities in multi-field turbulence converge remarkably faster than the single-field case, which requires thousands of modes to converge., Comment: 11 pages, 4 figures

The scalar spin chirality, which characterizes the fundamental unit of noncoplanar spin structures, plays an important role in rich chiral physics of magnetic materials. In particular, the intensive research efforts over the past two decades have demonstrated that the scalar spin chirality is the source of various novel Hall transports in solid-state systems, offering a primary route to bring about chiral phenomena in condensed matter physics. However, in all of the previous studies, the scalar spin chirality has been given as a stationary background, serving only a passive role in the transport properties of materials. It remains an open question whether or not the scalar spin chirality itself can exhibit a Hall-type transport. In this work, we show that the answer is yes: The scalar spin chirality is Hall-transported in Kagome ferromagnets and antiferromagnets under an external bias, engendering a phenomenon which we dub the scalar spin chirality Hall effect. Notably, this effect is present even in the absence of any spin-orbit coupling. The analytical theory for the scalar spin chirality Hall effect is corroborated by atomistic spin simulations. Our findings call for the need to lift the conventional assumption that the scalar spin chirality is a passive background in order to discover the active roles of the scalar spin chirality in transport properties., Comment: 9 pages, 4 figures

Adam is one of the most popular optimization algorithms in deep learning. However, it is known that Adam does not converge in theory unless choosing a hyperparameter, i.e., $\beta_2$, in a problem-dependent manner. There have been many attempts to fix the non-convergence (e.g., AMSGrad), but they require an impractical assumption that the gradient noise is uniformly bounded. In this paper, we propose a new adaptive gradient method named ADOPT, which achieves the optimal convergence rate of $\mathcal{O} ( 1 / \sqrt{T} )$ with any choice of $\beta_2$ without depending on the bounded noise assumption. ADOPT addresses the non-convergence issue of Adam by removing the current gradient from the second moment estimate and changing the order of the momentum update and the normalization by the second moment estimate. We also conduct intensive numerical experiments, and verify that our ADOPT achieves superior results compared to Adam and its variants across a wide range of tasks, including image classification, generative modeling, natural language processing, and deep reinforcement learning. The implementation is available at https://github.com/iShohei220/adopt., Comment: Accepted at Neural Information Processing Systems (NeurIPS 2024)

We further develop the BPS/CFT correspondence between quiver W-algebras/$qq$-characters and partition functions of gauge origami. We introduce $qq$-characters associated with multi-dimensional partitions with nontrivial boundary conditions which we call Donaldson--Thomas (DT) $qq$-characters. They are operator versions of the equivariant DT vertices of toric Calabi--Yau three and four-folds. Moreover, we revisit the construction of the D8 $qq$-characters with no boundary conditions and give a quantum algebraic derivation of the sign rules of the magnificent four partition function. We also show that under the proper sign rules, the D6 and D8 $qq$-characters with no boundary conditions all commute with each other and discuss its physical interpretation., Comment: 66+13 pages

This study examines the educational effect of the Academic Support Center at Kogakuin University. Following the initial assessment, it was suggested that group bias had led to an underestimation of the Center's true impact. To address this issue, the authors applied the theory of causal inference. By using T-learner, the conditional average treatment effect (CATE) of the Center's face-to-face (F2F) personal assistance program was evaluated. Extending T-learner, the authors produced a new CATE function that depends on the number of treatments (F2F sessions) and used the estimated function to predict the CATE performance of F2F assistance., Comment: 12 pages, 7 figures

Large-scale pre-trained models (PTMs) provide remarkable zero-shot classification capability covering a wide variety of object classes. However, practical applications do not always require the classification of all kinds of objects, and leaving the model capable of recognizing unnecessary classes not only degrades overall accuracy but also leads to operational disadvantages. To mitigate this issue, we explore the selective forgetting problem for PTMs, where the task is to make the model unable to recognize only the specified classes while maintaining accuracy for the rest. All the existing methods assume "white-box" settings, where model information such as architectures, parameters, and gradients is available for training. However, PTMs are often "black-box," where information on such models is unavailable for commercial reasons or social responsibilities. In this paper, we address a novel problem of selective forgetting for black-box models, named Black-Box Forgetting, and propose an approach to the problem. Given that information on the model is unavailable, we optimize the input prompt to decrease the accuracy of specified classes through derivative-free optimization. To avoid difficult high-dimensional optimization while ensuring high forgetting performance, we propose Latent Context Sharing, which introduces common low-dimensional latent components among multiple tokens for the prompt. Experiments on four standard benchmark datasets demonstrate the superiority of our method with reasonable baselines. The code is available at https://github.com/yusukekwn/Black-Box-Forgetting., Comment: NeurIPS 2024

Chirality is ubiquitous in nature and manifests in a wide range of phenomena including chemical reactions, biological processes, and quantum transport of electrons. In quantum materials, the chirality of fermions, given by the relative directions between the electron spin and momentum, is connected to the band topology of electronic states. Here, we show that in structurally chiral materials like CoSi, the orbital angular momentum (OAM) serves as the main driver of a nontrivial band topology in this new class of unconventional topological semimetals, even when spin-orbit coupling is negligible. A nontrivial orbital-momentum locking of multifold chiral fermions in the bulk leads to a pronounced OAM texture of the helicoid Fermi arcs at the surface. Our findings highlight the pivotal role of the orbital degree of freedom for the chirality and topology of electron states, in general, and pave the way towards the application of topological chiral semimetals in orbitronic devices.

Altermagnets are magnetic materials with antiferromagnetic spin ordering but exhibit ferromagnetic properties. Understanding the microscopic origin of the latter is a central problem. Ferromagnet-like properties such as the anomalous Hall effect are linked with weak ferromagnetism, whose microscopic origin in altermagnets remains unclear however. We show theoretically that the alternating $g$-tensor anisotropy in altermagnets can induce weak ferromagnetism even when the Dzyaloshinskii-Moriya interaction is forbidden. We demonstrate this mechanism for both collinear and noncollinear spin altermagnets and explain various properties of the weak ferromagnetism in altermagnets. Our findings provide insights for characterizing and manipulating magnetic configurations in altermagnets.

A measure of asymmetry is a quantification method that allows for the comparison of categorical evaluations before and after treatment effects or among different target populations, irrespective of sample size. We focus on square contingency tables that summarize survey results between two time points or cohorts, represented by the same categorical variables. We propose a measure to evaluate the degree of departure from a symmetry model using cosine similarity. This proposal is based on the Fisher-Rao distance, allowing asymmetry to be interpreted as a geodesic distance between two distributions. Various measures of asymmetry have been proposed, but visualizing the relationship of these quantification methods on a two-dimensional plane demonstrates that the proposed measure provides the geometrically simplest and most natural quantification. Moreover, the visualized figure indicates that the proposed method for measuring departures from symmetry is less affected by very few cells with extreme asymmetry. A simulation study shows that for square contingency tables with an underlying asymmetry model, our method can directly extract and quantify only the asymmetric structure of the model, and can more sensitively detect departures from symmetry than divergence-type measures.

Large Language Models (LLMs) have gained significant attention in on-device applications due to their remarkable performance across real-world tasks. However, on-device LLMs often suffer from suboptimal performance due to hardware limitations. A promising solution to this challenge is cascading a weaker local (on-device) LLM with a more powerful server LLM. While existing research on LLM cascade primarily optimizes the performance-cost trade-off, real-world applications impose additional requirements, such as privacy preservation, which remain largely unaddressed. In this work, we move beyond existing confidence- and logit-based LLM cascade methods and propose $\mathbf{P^{3}Defer}$, a novel Chain-of-Thought (CoT)-enhanced \textbf{p}olicy learning framework for \textbf{p}rivacy-\textbf{p}reserved \textbf{defer}ral decision-making. Our approach effectively improves cascade efficiency while mitigating privacy risks. Extensive experiments on three benchmark datasets demonstrate the effectiveness and superiority of $\mathbf{P^{3}Defer}$ over existing methods.

Molecular dynamics (MD) simulations were conducted using the generalized replica exchange method (gREM) on the 4-cyano-4$^{\prime}$-$n$-alkylbiphenyl ($n$CB) system with $n=5$, 6, 7, and 8, which exhibits a nematic-isotropic (NI) phase transition. Sampling near the phase transition temperature in systems undergoing first-order phase transitions, such as the NI phase transition, is demanding due to the substantial energy gap between the two phases. To address this, gREM, specifically designed for first-order phase transitions, was utilized to enhance sampling near the NI phase transition temperature. Free-energy calculations based on the energy representation (ER) theory were employed to characterize the NI phase transition. ER evaluates the insertion free energy of $n$CB molecule for both nematic and isotropic phases, revealing a change in the temperature dependence across the NI phase transition. Further decomposition into intermolecular interaction energetic and entropic terms shows quantitatively the balance between these contributions at the NI phase transition temperature., Comment: 9 pages, 5 figures for main text, 8 pages for Supplementary Material