Search

While recent text-to-image models can generate photorealistic images from text prompts that reflect detailed instructions, they still face significant challenges in accurately rendering words in the image. In this paper, we propose to retouch erroneous text renderings in the post-processing pipeline. Our approach, called Type-R, identifies typographical errors in the generated image, erases the erroneous text, regenerates text boxes for missing words, and finally corrects typos in the rendered words. Through extensive experiments, we show that Type-R, in combination with the latest text-to-image models such as Stable Diffusion or Flux, achieves the highest text rendering accuracy while maintaining image quality and also outperforms text-focused generation baselines in terms of balancing text accuracy and image quality.

Materials with J-shaped stress-strain behavior under uniaxial stretching, where strength increases as deformation progresses, have been developed through various materials designs. On the other hand, polymer materials that progressively stiffen under bending remain unrealized. To address this gap, this study drew inspiration from membrane tensegrity structures, which achieve structural stability by balancing compressive forces in rods and tensile forces in membrane. Notably, some of these structures exhibit increased stiffness under bending. Using a multipolymer patterning technique, we developed a polymer film exhibiting membrane tensegrity-like properties that stiffens under bending. This effect results from membrane tension generated by rod protrusions and an increase in second moment of area at regions with maximum curvature.

The question that drives this research is: "How to discover the number of respondents that are necessary to validate items of a questionnaire as actually essential to reach the questionnaire's proposal?" Among the efforts in this subject, \cite{Lawshe1975, Wilson2012, Ayre_CVR_2014} approached this issue by proposing and refining the Content Validation Ratio (CVR) that looks to identify items that are actually essentials. Despite their contribution, these studies do not check if an item validated as "essential" should be also validated as "not essential" by the same sample, which should be a paradox. Another issue is the assignment a probability equal a 50\% to a item be randomly checked by a respondent as essential, despite an evaluator has three options to choose. Our proposal faces these issues, making it possible to verify if a paradoxical situation occurs, and being more precise in recommending whether an item should either be retained or discarded from a questionnaire., Comment: 17 pages, 1 figure

Recent advancements in foundation models show promising capability in graphic design generation. Several studies have started employing Large Multimodal Models (LMMs) to evaluate graphic designs, assuming that LMMs can properly assess their quality, but it is unclear if the evaluation is reliable. One way to evaluate the quality of graphic design is to assess whether the design adheres to fundamental graphic design principles, which are the designer's common practice. In this paper, we compare the behavior of GPT-based evaluation and heuristic evaluation based on design principles using human annotations collected from 60 subjects. Our experiments reveal that, while GPTs cannot distinguish small details, they have a reasonably good correlation with human annotation and exhibit a similar tendency to heuristic metrics based on design principles, suggesting that they are indeed capable of assessing the quality of graphic design. Our dataset is available at https://cyberagentailab.github.io/Graphic-design-evaluation ., Comment: Accepted to SIGGRAPH Asia 2024 (Technical Communications Track)

We study the energetics of bursting activity in a magnetar under the `Dzhanibekov effect,' described by the classical mechanics of rigid bodies. The magnetar's rotation axis suddenly flips due to this effect, resulting in a sudden rise of the Euler force. The outer layer of magnetar, called the crust, can plastically flow and eventually break due to the Euler force. Then, the degenerate electrons burst open from the braking region like a balloon burst. We find that the pair plasma formation can be ignited inside the crust, and its radiation can be similar to the observed bursts from soft gamma-ray repeaters. At the beginning of this bursting phenomenon, the self-discharge effect of the electron streaming may induce a strong resistive electric field, accelerating the ions. We also find that the maximum energy of the accelerated ions possibly reaches $\sim$ZeV within a timescale of $\sim1$~ps. If $\gtrsim2$~\% of magnetars in nearby galaxies are undergoing this scenario, the observed cosmic-ray flux at $>100$~EeV can be explained., Comment: submitted to PTEP (13 pages of the main text with a single column, 4 figures, 22 pages in total)

Human physical function is governed by self-efficacy, the belief in one's motor capacity. In chronic pain patients, this capacity may remain reduced long after the damage causing the pain has been cured. Chronic pain alters body schema, affecting how patients perceive the dimension and pose of their bodies. We exploit this deficit using robotic manipulation technology and augmented sensory stimuli through virtual reality technology. We propose a sensory stimuli manipulation method aimed at modifying body schema to restore lost self-efficacy.

Parallelization of Greedy Best First Search (GBFS) has been difficult because straightforward parallelization can result in search behavior which differs significantly from sequential GBFS, exploring states which would not be explored by sequential GBFS with any tie-breaking strategy. Recent work has proposed a class of parallel GBFS algorithms which constrains search to exploration of the Bench Transition System (BTS), which is the set of states that can be expanded by GBFS under some tie-breaking policy. However, enforcing this constraint is costly, as such BTS-constrained algorithms are forced to spend much of the time waiting so that only states which are guaranteed to be in the BTS are expanded. We propose an improvement to parallel search which decouples state generation and state evaluation and significantly improves state evaluation rate, resulting in better search performance., Comment: In Proceedings of ICAPS-2024 Workshop on Heuristics and Search for Domain-Independent Planning (HSDIP-24) https://icaps24.icaps-conference.org/program/workshops/hsdip/

The typical material bismuth shows a several-fold change in electrical resistivity at 4 K over a 400 {\AA} film thickness period. In 2021, we discovered a novel size effect with a period of 25 {\AA} on strongly correlated compound CaRuO_3. The change in film thickness of only one nanometer leads to an increase in electrical resistivity at 4 K by a factor of several billion. However, the excitation energy of 2.4 eV on insulating CaRuO_3 is too large to explain the mechanism by conventional quantum well. In this study, we clarify that the increases in electrical resistivity are accompanied by lattice expansion and caused by the Mott transitions. We measured in-plane X-ray diffraction on CaRuO_3 films and found a 24.8 {\AA} periodic thickness oscillation of the lattice spacing d_{(004)}. We determined the nesting vector from the Fermi surface and revealed the spin density wave with a period equivalent to the size effect. We concluded that the antiferromagnetic correlation appearing in the boundary conditions triggers the periodic Mott insulators depending on the film thickness. We refer to the remarkable size effect stemming from strong electron correlation as the 'Mott size effect'., Comment: 28 pages, 15 figures

The afterglows of gamma-ray bursts are non-thermal electron synchrotron emissions from relativistic shocks. The origin of strong magnetic field in the emission region remains elusive, and two field amplification mechanisms via the plasma kinetic and magnetohydrodynamic instabilities have been discussed. The polarimetric observations are a powerful probe to distinguish these two mechanisms. So far, most theoretical works have focused on the former mechanism and constructed afterglow polarization models with microscopic-scale turbulence whose coherence length is much smaller than the thickness of the blast wave. In this work, focusing on the latter mechanism, we utilize our semi-analytic model of the synchrotron polarization with large-scale turbulence whose coherence length is comparable to the thickness of the blast wave to investigate the effect of magnetic field anisotropy and the observer viewing angle. We find that the polarization in our large-scale turbulence model can exhibit both behaviors characteristic of the microscopic-scale turbulence model and those not seen in the microscopic-scale model. Then we find that the large-scale model could explain all the polarimetric observational data to date that seem to be forward shock emission. We also examine the effect of ordered-field component, and find that polarization degree and polarization angle constant in time are realized only when the energy density ratio of the ordered and fluctuated components is $\gtrsim 50$. In this case, however, the polarization degree is much higher than the observed values., Comment: Submitted to ApJ, comments welcome

This paper will examine the cognitive processes involved in palpation in order to develop an appropriate remote palpation system. In a conventional remote palpation system, the tactile condition of the patient is conveyed to the doctors using a force feedback system. A clarification of the cognitive process during palpation suggests that the purpose of palpation is to formulate a clear idea about the patient's medical problems using the tactile sensation as a trigger to combine the results of other assessments, past experience and memory, and patient reactions to the doctor's touch. This is in contrast to the objective of acquiring the detailed tactile condition of the affected body part. In order to demonstrate this purpose, we will describe the two significant signal pathways for the perception of tactile sensation, both in doctors and patients. The perception of doctors progresses as the result of active touch to the affected part, thereby implying that the simultaneous stimulation of kinaesthetic and tactile sensation is necessary. Conversely, the tactile sensation experienced by patients is the result of passive touch, which evokes a more subjective and emotional response. Patients both explicitly and implicitly perceive the stimulation, and doctors use these perceptions as reactions of the pain to the doctors' touch. This paper proposes the fundamental concept of a remote palpation system, ``Palpation Reality beyond Real'', to achieve the purpose of palpation. Palpation reality implies a system in which the whole cognitive process progresses at the same level or better than palpation in the standard examination, rather than presenting the real tactile sensation.

Automatic generation of graphic designs has recently received considerable attention. However, the state-of-the-art approaches are complex and rely on proprietary datasets, which creates reproducibility barriers. In this paper, we propose an open framework for automatic graphic design called OpenCOLE, where we build a modified version of the pioneering COLE and train our model exclusively on publicly available datasets. Based on GPT4V evaluations, our model shows promising performance comparable to the original COLE. We release the pipeline and training results to encourage open development., Comment: To appear as an extended abstract (EA) in Workshop on Graphic Design Understanding and Generation (in CVPR2024), code: https://github.com/CyberAgentAILab/OpenCOLE

In this research, multi-physics topology optimization is employed to achieve the detailed design of a lightweight porous linear actuation mechanism that harnesses energy through laser activation. A multiscale topology optimization methodology is introduced for micro- and macroscale design, considering energy dissipation via heat convection and radiation. This investigation meticulously considers the impact of heat dissipation mechanisms, including thermal conduction, convection, and radiation. Through various numerical cases, we systematically explore the influence of micro-scale considerations on porous design and understand the effects on the topology optimization process by incorporating various microstructural systems. The results demonstrate that porous actuator designs exhibit superior performance compared to solid actuator designs. This study contributes to advancing the understanding of multiscale effects in topology optimization, paving the way for more efficient and lightweight designs in the field of laser-activated porous actuators.

From 1998 to 2017, neutron monitors located at an altitude of 4300 m on the Tibetan plateau detected 127 long-duration bursts of high-energy radiation in association with thunderclouds. These bursts typically lasted for 10 to 40 minutes, and 89\% of them occurred between 10:00 and 24:00 local time. They were also found to be more likely to occur at night, especially during 18:00$-$06:00 local time period. The observed diurnal and seasonal variations in burst frequency were consistent with the frequencies of lightning and precipitation on the Tibetan plateau. Based on 19 years of data, the present study suggests that an annual variation in burst frequency has a periodicity of $\sim$16 years and a lag of $\sim$3 years relative to solar activity., Comment: 28 pages, accepted for publication in Progress in Earth and Planetary Science

The Fermi and eROSITA bubbles, large diffuse structures in our Galaxy, can be the by-products of the steady star formation activity. To simultaneously explain the star formation history of the Milky Way and the metallicity of $\sim$ Z$_\odot$ at the Galactic disk, a steady Galactic wind driven by cosmic-rays is required. For tenuous gases with a density of $\lesssim$10$^{-3}$ cm$^{-3}$, the cosmic-ray heating dominates over radiative cooling, and the gas can maintain the virial temperature of $\sim$0.3 keV ideal for escape from the Galactic system as the wind. A part of the wind falls back onto the disk like a galactic fountain flow. We model the wind dynamics according to the Galactic evolution scenario and find that the scale height and surface brightness of the X-ray and the hadronic gamma-ray emissions from such fountain flow region can be consistent with the observed properties of the Fermi and eROSITA bubbles. This implies that the bubbles are persistent structures of the Milky Way existing over (at least) the last $\sim$1 Gyr, rather than evanescent structures formed by non-trivial, $\sim$10 Myr past Galactic Center transient activities., Comment: 13 pages, 9 figures, accepted for publication in ApJ

In this paper, we demonstrate a total disentanglement of font images. Total disentanglement is a neural network-based method for decomposing each font image nonlinearly and completely into its style and content (i.e., character class) features. It uses a simple but careful training procedure to extract the common style feature from all `A'-`Z' images in the same font and the common content feature from all `A' (or another class) images in different fonts. These disentangled features guarantee the reconstruction of the original font image. Various experiments have been conducted to understand the performance of total disentanglement. First, it is demonstrated that total disentanglement is achievable with very high accuracy; this is experimental proof of the long-standing open question, ``Does `A'-ness exist?'' Hofstadter (1985). Second, it is demonstrated that the disentangled features produced by total disentanglement apply to a variety of tasks, including font recognition, character recognition, and one-shot font image generation.

Magnetars, which are neutron stars with strong magnetic fields, exhibit occasional bursting activities. The shape of a magnetar is not perfectly spherical due to the Lorentz force exerted by its strong magnetic fields and is described as a triaxial body. We study the unstable free precession in a triaxial magnetar; one of the principal axes undergoes an upside-down flip. This flip is known as the Dzhanibekov effect. We find that during the flip, the Euler force can suddenly disturb the force balance on the surface layer of the magnetar, potentially leading to plastic flow of the layer. This, in turn, may trigger different forms of magnetar activity, such as the emission of the bursts and/or of gravitational waves., Comment: 8 pages, 4 figures, accepted for publication in ApJ

Image cropping is essential in image editing for obtaining a compositionally enhanced image. In display media, image cropping is a prospective technique for automatically creating media content. However, image cropping for media contents is often required to satisfy various constraints, such as an aspect ratio and blank regions for placing texts or objects. We call this problem image cropping under design constraints. To achieve image cropping under design constraints, we propose a score function-based approach, which computes scores for cropped results whether aesthetically plausible and satisfies design constraints. We explore two derived approaches, a proposal-based approach, and a heatmap-based approach, and we construct a dataset for evaluating the performance of the proposed approaches on image cropping under design constraints. In experiments, we demonstrate that the proposed approaches outperform a baseline, and we observe that the proposal-based approach is better than the heatmap-based approach under the same computation cost, but the heatmap-based approach leads to better scores by increasing computation cost. The experimental results indicate that balancing aesthetically plausible regions and satisfying design constraints is not a trivial problem and requires sensitive balance, and both proposed approaches are reasonable alternatives., Comment: ACMMM Asia accepted