Placement is a crucial and time-intensive step in the modern Electronic Design Automation (EDA) chip design process, involving the allocation of millions of modules on a chip canvas. Previous studies have demonstrated the efficacy of machine learning methods, especially reinforcement learning (RL) in chip placement. However, existing RL-based methods still face challenges, including lengthy placement times due to placing only a single module at each step, and a lack of generalization capability. In response to these challenges, we propose a Graph Convolutional Network (GCN)-based diffusion model called GraphDiffusion. Our model presents a novel approach for chip placement by representing the positioning of movable nodes as a conditional denoising diffusion process. The GCN is utilized to extract feature information from the circuit, which then serves as an embedding to guide the diffusion model in generating the chip layout. We exploit the diffusion model’s capabilities to improve layout quality and generalizability. We conduct extensive experiments on the ISPD benchmark. The results demonstrate that our approach outperforms advanced placement methods while enhancing the transferability of the layout model. Upon training with expert data, GraphDiffusion can interpret the input circuit netlists and generate state-of-the-art chip layouts.

Recent advancements in large language models (LLMs) underscore the need for more comprehensive evaluation methods to accurately assess their reasoning capabilities. Existing benchmarks are often domain-specific and thus cannot fully capture an LLM’s general reasoning potential. To address this limitation, we introduce the **Knowledge Orthogonal Reasoning Gymnasium (KORGym)**, a dynamic evaluation platform inspired by KOR-Bench and Gymnasium. KORGym offers over fifty games in either textual or visual formats and supports interactive, multi-turn assessments with reinforcement learning scenarios. Using KORGym, we conduct extensive experiments on 19 LLMs and 8 VLMs, revealing consistent reasoning patterns within model families and demonstrating the superior performance of closed-source models. Further analysis examines the effects of modality, reasoning strategies, reinforcement learning techniques, and response length on model performance. We expect KORGym to become a valuable resource for advancing LLM reasoning research and developing evaluation methodologies suited to complex, interactive environments.

Long-tailed semi-supervised learning poses a significant challenge in training models with limited labeled data exhibiting a long-tailed label distribution. Current state-of-the-art LTSSL approaches heavily rely on high-quality pseudo-labels for large-scale unlabeled data. However, these methods often neglect the impact of representations learned by the neural network and struggle with real-world unlabeled data, which typically follows a different distribution than labeled data. This paper introduces a novel probabilistic framework that unifies various recent proposals in long-tail learning. Our framework derives the class-balanced contrastive loss through Gaussian kernel density estimation. We introduce a continuous contrastive learning method, CCL, extending our framework to unlabeled data using *reliable* and *smoothed* pseudo-labels. By progressively estimating the underlying label distribution and optimizing its alignment with model predictions, we tackle the diverse distribution of unlabeled data in real-world scenarios. Extensive experiments across multiple datasets with varying unlabeled data distributions demonstrate that CCL consistently outperforms prior state-of-the-art methods, achieving over 4% improvement on the ImageNet-127 dataset. The supplementary material includes the source code for reproducibility.

The emotion regulation mechanism of mindfulness plays an important role in the stress reduction effect. Many researchers in the fields of cognitive psychology and cognitive neuroscience have attempted to elucidate this mechanism by documenting the cognitive processes that occur and the neural activities that characterize each process. However, previous findings have not revealed the mechanism of information propagation in the brain that achieves emotion regulation during mindfulness. In this study, we constructed a functional brain model based on its anatomical network structure and a computational model representing the propagation of information between brain regions. We then examined the effects of mindfulness meditation on information propagation in the brain using simulations of changes in the activity of each region. These simulations of changes represent the degree of processing resource allocation to the neural activity via changes in the weights of each region's output. As a result of the simulations, we reveal how the neural activity characteristic of emotion regulation in mindfulness, which has been reported in previous studies, is realized in the brain. Mindfulness meditation increases the weight of the output from each region of the thalamus and sensory cortex, which processes sensory stimuli from the external world. This sensory information activates the insula and anterior cingulate cortex (ACC). The orbitofrontal cortex and dorsolateral prefrontal cortex inhibit amygdala activity (i.e., top-down emotion regulation). However, when mindfulness meditation dominates bottom-up processing via sensory stimuli from the external world, amygdala activity increases through the insula and ACC activation.

Knee osteoarthritis (KOA) has been developed to be one of the most popular chronic joint disease in elder. Muscle shrink or force reduction of lower limb muscles such as gastrocnemius and quadriceps is a pathogenic factor about KOA. Therefore, imaging techniques such as MRI and CT for observing muscles morphological difference become popular in diagnosing KOA in clinic. However, the image information in different populations may vary according to age, weight, sex, and may be influenced by the subjective judgment according to doctor's experience. Viscoelasticity is a quantitative indicator for assessing the biomechanical properties of muscle, but there's lack of research to describe the relationship between viscoelasticity of lower limb muscles and KOA. The purpose of this study is to discuss the difference of viscoelasticity during lower limb muscles paining. We designed an experiment called muscle perfusion to simulate the muscle pain of KOA patients, and measured the lower limb muscles' viscoelasticity using shear-wave dispersion ultra-sound vibrometry (SDUV). Besides, we detected the viscoelasticity changes of a volunteer who had lower limb pain (which was similar to the symptom of KOA) during Fu's Subcutaneous Needling (FSN) treatment. Results show that during FSN treatment as well as perfusion, the elasticity and viscosity of lower limb muscles changed significantly, showing that viscoelasticity was a potential quantitative index to diagnose and qualify the disease of KOA.

This paper presents a system for producing multiperspective panoramas for long scenes from dense collections of images. Multi-perspective panoramas are generated by combining different perspectives, including original and novel perspectives. The latter are rendered by our synthesis algorithm, which uses estimated depth information to remove distortions caused by depth parallax. Then, different perspectives are properly combined without noticeable visual artifacts caused by both color inconsistencies and structural misalignments.