The composition dependence of gravitational constant $G$ is measured at the millimeter scale to test the weak equivalence principle, which may be violated at short range through new Yukawa interactions such as the dilaton exchange force. A torsion balance on a turning table with two identical tungsten targets surrounded by two different attractor materials (copper and aluminum) is used to measure gravitational torque by means of digital measurements of a position sensor. Values of the ratios $\tilde{G}_{Al-W}/\tilde{G}_{Cu-W} -1$ and $\tilde{G}_{Cu-W}/G_{N} -1$ were $(0.9 \pm 1.1_{\mathrm{sta}} \pm 4.8_{\mathrm{sys}}) \times 10^{-2}$ and $ (0.2 \pm 0.9_{\mathrm{sta}} \pm 2.1_{\mathrm{sys}}) \times 10^{-2}$ , respectively; these were obtained at a center to center separation of 1.7 cm and surface to surface separation of 4.5 mm between target and attractor, which is consistent with the universality of $G$. A weak equivalence principle (WEP) violation parameter of $\eta_{Al-Cu}(r\sim 1\: \mathrm{cm})=(0.9 \pm 1.1_{\mathrm{sta}} \pm 4.9_{\mathrm{sys}}) \times 10^{-2} $ at the shortest range of around 1 cm was also obtained.

K. Ninomiya T. Akiyama M. Hata M. Hatori T. Iguri Y. Ikeda S. Inaba H. Kawamura R. Kishi H. Murakami Y. Nakaya H. Nishio N. Ogawa J. Onishi S. Saiba T. Sakuta S. Tanaka R. Tanuma Y. Totsuka R. Tsutsui K. Watanabe J. Murata

This paper presents LEMONADE, a large-scale conflict event dataset comprising 39,786 events across 20 languages and 171 countries, with extensive coverage of region-specific entities. LEMONADE is based on a partially reannotated subset of the Armed Conflict Location & Event Data (ACLED), which has documented global conflict events for over a decade. To address the challenge of aggregating multilingual sources for global event analysis, we introduce abstractive event extraction (AEE) and its subtask, abstractive entity linking (AEL). Unlike conventional span-based event extraction, our approach detects event arguments and entities through holistic document understanding and normalizes them across the multilingual dataset. We evaluate various large language models (LLMs) on these tasks, adapt existing zero-shot event extraction systems, and benchmark supervised models. Additionally, we introduce ZEST, a novel zero-shot retrieval-based system for AEL. Our best zero-shot system achieves an end-to-end F1 score of 58.3%, with LLMs outperforming specialized event extraction models such as GoLLIE. For entity linking, ZEST achieves an F1 score of 45.7%, significantly surpassing OneNet, a state-of-the-art zero-shot baseline that achieves only 23.7%. However, these zero-shot results lag behind the best supervised systems by 20.1% and 37.0% in the end-to-end and AEL tasks, respectively, highlighting the need for further research.

Sina J. Semnani Pingyue Zhang Wanyue Zhai Haozhuo Li Ryan Beauchamp Trey Billing Katayoun Kishi Manling Li Monica S. Lam

We construct some families of quadratic fields whose class numbers are divisible by $3.$ The main tools used are a trinomial introduced by Kishi and a parametrization of Kishi and Miyake of a family of quadratic fields whose class numbers are divisible by $3.$ At the end we compute class number of these fields for some small values and verify our results.

Azizul Hoque Kalyan Chakraborty

The $^{11}$B($^3$He$, t$), $^{11}$B($d, d'$), and $^{11}$B($p, p'$) reactions were measured at forward scattering angles including $0^\circ$ to study the isovector and isoscalar spin-flip M1 strengths in $^{11}$B. The measured $^{11}$B($^3$He$, t$) cross sections were compared with the results of the distorted-wave impulse-approximation (DWIA) calculation, and the Gamow-Teller (GT) strengths for low-lying states in $^{11}$C were determined. The GT strengths were converted to the isovector spin-flip M1 strengths using the isobaric analog relations under the assumption of the isospin symmetry. The isoscalar spin-flip M1 strengths were obtained from the ($d, d'$) analysis by assuming that the shape of the collective transition form factor with the same ${\Delta}J^\pi$ is similar in the $^{11}$B($d, d'$) and $^{12}$C($d, d'$) reactions. The obtained isovector and isoscalar strengths were used in the DWIA calculations for the $^{11}$B($p, p'$) reaction. The DWIA calculation reasonably well explains the present $^{11}$B($p, p'$) result. However, the calculated cross section for the 8.92-MeV 3/2$^-_2$ state was significantly smaller than the experimental values. The transition strengths obtained in the shell-model calculations were found to be 20-50% larger than the experimental strengths. The transition strengths for the neutrino induced reactions were estimated by using the isovector and isoscalar spin-flip M1 strengths. The present results are quantitatively in agreement with the theoretical estimation discussing the axial isoscalar coupling in the neutrino scattering process, and are useful in the measurement of the stellar neutrinos using the neutral- and charged-current reactions on $^{11}$B.

T. Kawabata H. Akimune H. Fujimura H. Fujita Y. Fujita M. Fujiwara K. Hara K. Y. Hara K. Hatanaka T. Ishikawa M. Itoh J. Kamiya S. Kishi M. Nakamura K. Nakanishi T. Noro H. Sakaguchi Y. Shimbara H. Takeda A. Tamii S. Terashima H. Toyokawa M. Uchida H. Ueno T. Wakasa Y. Yasuda H. P. Yoshida M. Yosoi

We extend the Kubo-Kishi theorem concerning the charge susceptibility of the Hubbard model in the following way: (i) The electron-photon interaction is taken into account. (ii) Not only on-site but also general Coulomb repulsion is considered.

Tadahiro Miyao

For a given odd integer $n>1$, we provide some families of imaginary quadratic number fields of the form $\mathbb{Q}(\sqrt{x^2-t^n})$ whose ideal class group has a subgroup isomorphic to $\mathbb{Z}/n\mathbb{Z}$.

Kalyan Chakraborty Azizul Hoque Yasuhiro Kishi Prem Prakash Pandey

Dielectric responces of the one-dimentional electron system is investigated numerically. We treat an interacting one-dimentional spinless fermion model with disorder by using the Density Matrix Renormalization Group(DMRG) method which is extended for nonuniform systems. We apply an electric field $E$ to the system and calculate dielectric responces. Dielectric responce of the Mott insulator and the Anderson insulator are calculated respectively. Steplike behaviors in $P-E$ curve are obtained which corresponds to breakdown of the insulating behavior. For the Mott insulator, the steps originate from generation of kink-pairs. For the Anderson insulator on the other hand, the origin of the steps is a crossing of the localized one particle energy levels. We also treat random systems with interaction. From one parameter scaling analysis of the susceptibility $\chi$, the metal-insulator transition in attractively interacting region is confirmed and a phase diagram of the random spinless fermion model is obtained.

Masato Kishi Yasuhiro Hatsugai

Herein,we numerically examine the relative dispersion of Lagrangian particle pairs in two-dimensional inverse energy-cascade turbulence. Behind the Richardson-Obukhov $t^3$ law of relative separation, we discover that the second-order moment of the relative velocity have a temporal scaling exponent different from the prediction based on the Kolmogorov's phenomenology. The results also indicate that time evolution of the probability distribution function of the relative velocity is self-similar. The findings are obtained by enforcing Richardson-Obukhov law either by considering a special initial separation or by conditional sampling. In particular, we demonstrate that the conditional sampling removes the initial-separation dependence of the statistics of the separation and relative velocity. Furthermore, we demonstrate that the conditional statistics are robust with respect to the change in the parameters involved, and that the number of the removed pairs from the sampling decreases when the Reynolds number increases. We also discuss the insights gained as a result of conditional sampling.

Tatsuro Kishi Takeshi Matsumoto Sadayoshi Toh

Perturbative GAN, which replaces convolution layers of existing convolutional GANs (DCGAN, WGAN-GP, BIGGAN, etc.) with perturbation layers that adds a fixed noise mask, is proposed. Compared with the convolu-tional GANs, the number of parameters to be trained is smaller, the convergence of training is faster, the incep-tion score of generated images is higher, and the overall training cost is reduced. Algorithmic generation of the noise masks is also proposed, with which the training, as well as the generation, can be boosted with hardware acceleration. Perturbative GAN is evaluated using con-ventional datasets (CIFAR10, LSUN, ImageNet), both in the cases when a perturbation layer is adopted only for Generators and when it is introduced to both Generator and Discriminator.

Yuma Kishi Tsutomu Ikegami Shin-ichi O'uchi Ryousei Takano Wakana Nogami Tomohiro Kudoh

Graph kernels are often used in bioinformatics and network applications to measure the similarity between graphs; therefore, they may be used to construct efficient graph classifiers. Many graph kernels have been developed thus far, but to the best of our knowledge there is no existing graph kernel that considers all subgraphs to measure similarity. We propose a novel graph kernel that applies a quantum computer to measure the graph similarity taking all subgraphs into account by fully exploiting the power of quantum superposition to encode every subgraph into a feature. For the construction of the quantum kernel, we develop an efficient protocol that removes the index information of subgraphs encoded in the quantum state. We also prove that the quantum computer requires less query complexity to construct the feature vector than the classical sampler used to approximate the same vector. A detailed numerical simulation of a bioinformatics problem is presented to demonstrate that, in many cases, the proposed quantum kernel achieves better classification accuracy than existing graph kernels.

Kaito Kishi Takahiko Satoh Rudy Raymond Naoki Yamamoto Yasubumi Sakakibara