Chen’s research while affiliated with Southwest University and other places

Open cluster (OC) groups are collections of spatially close OCs originating from the same giant molecular cloud. The formation mechanism of the OC groups remains unclear due to limited sample size and data precision. Recent advances in astrometric data from the mission have provided an unprecedented opportunity to study OC groups in detail. We aim to extend the sample of OC groups and explore the formation and evolution mechanisms of the newly identified primordial OC groups. We focus on the impact of stellar feedback events occurring near the OC groups and their role in triggering star formation within these groups. We identified the OC group using data by analyzing the close correlations in three-dimensional (3D) spatial, 3D velocity, and age. We conducted N-body simulations to trace the dynamical evolution of these groups and obtained the birthplace of OCs. A region where supernova (SN) explosions most likely occurred was predicted around the birthplaces of OC groups, based on the correlation between OC ages and their separation from the possible SN explosion sites. We also traced the orbits of pulsars (PSRs) using the Galactic potential model to probe their association with predicted SN explosion regions. We report the detection of four OC groups. The member OCs within each group are spatially proximate and exhibit similar velocities. The age spread of these OC groups is within 30 Myr, consistent with the duration of continuous star formation events. Dynamical simulations show that these OC groups gradually disperse over time, eventually evolving into independent OCs. The inference can be made that there exist specific regions surrounding Group 1 and Group 2 where the occurrence of SN explosions is highly probable. The strong correlations between OC ages and their separation from predicted SN explosion sites reveal a notable age gradient outward from the SN explosions. Additionally, we detected three PSRs near Group 1 and 26 PSRs near Group 2, whose birthplaces align with the predicted SN explosions regions. The member OCs within each OC group originate from the same molecular cloud, forming through a process of sequential star formation. We propose a star formation scenario in which multiple SN explosions triggered the formation of Group 1 and Group 2. Our results support the supernova-triggered star formation process and also reinforce the hierarchical star formation model, highlighting the multi-scale interactions that drive star and open cluster formation.

This paper introduces a data-driven algorithm for modeling and compensating shape deviations in additive manufacturing (AM), addressing challenges in geometric accuracy and batch production. While traditional methods, such as analytical models and metrology, laid the groundwork for geometric precision, they are often impractical for large-scale production. Recent advancements in machine learning (ML) have improved compensation precision, but issues remain in generalizing across complex geometries and adapting to position-dependent variations. We present a novel approach for powder bed fusion (PBF) processes, using GraphCompNet, which is a computational framework combining graph-based neural networks with a generative adversarial network (GAN)-inspired training process. By leveraging point cloud data and dynamic graph convolutional neural networks (DGCNNs), GraphCompNet models complex shapes and incorporates position-specific thermal and mechanical factors. A two-stage adversarial training procedure iteratively refines compensated designs via a compensator-predictor architecture, offering real-time feedback and optimization. Experimental validation across diverse shapes and positions shows the framework significantly improves compensation accuracy (35 to 65 percent) across the entire print space, adapting to position-dependent variations. This work advances the development of Digital Twin technology for AM, enabling scalable, real-time monitoring and compensation, and addressing critical gaps in AM process control. The proposed method supports high-precision, automated industrial-scale design and manufacturing systems.

pulposus cells; NLRP3, nucleotide-binding oligomerization domain, leucine-rich repeat and pyrin domain-containing 3; PTEN, phosphatase and tensin homolog deleted on chromosome ten; AKT, protein kinase B. Citation Zhou Y, Huang YY, Shang DY, et al. Advances in ethnopharmacology, phytochemistry and pharmacology of Erigerontis Herba and its Chinese medicine prescriptions. Abstract Erigerontis Herba (EH), the dried whole plant of Erigeron breviscapus, is well-known for circulating blood, activating meridians to alleviate pain, expelling wind, and clearing away cold. It has been extensively utilized in southern China for the treatment of stroke hemiplegia, chest stuffiness and pains, rheumatic arthralgia, headache, and toothache. This review focuses on the botany, ethnopharmacology, phytochemistry, pharmacology and toxicity of EH and its related prescriptions to offer new insights for prospective research of EH. Relevant information about EH was retrieved from ancient records and books, PubMed, China National Knowledge Infrastructure, Chinese Pharmacopoeia, Web of Science, Doctoral and Master's Theses, and various electronic databases. EH is a member of Compositae family and is mainly grown in southern China. Traditional Chinese medicine records that EH has the effects of circulating blood and removing blood stasis, expelling wind, and removing cold, as well as relieving rigidity of muscle and relieving pain. By now, nearly 200 ingredients have been characterized from EH, including flavonoids, caffeoyls, aromatic acids, coumarins, pentacyclic terpenoids, volatile oil and other compounds. EH extracts, EH related prescriptions (Dengzhan Xixin injection, Dengzhan Shengmai capsules, etc.) or compounds (scutellarin, scutellarein, etc.) possessed obvious therapeutic effects of ischemic stroke, cerebral hemorrhage, myocardial infarction, Alzheimer's disease, diabetes and its complications, gastric cancer, bone, and joint degenerative diseases. Scutellarin, the major active compound of EH, has been used as a quality marker. And no obvious toxicity of EH has been reported. According to its traditional applications, ethnopharmacology, phytochemistry, pharmacology, and toxicity, EH was applied as a valuable herb for clinical application in food and medicine fields. While several compounds have been shown to possess diverse biological activities, the underlying mechanisms of their actions remain elusive. To fully exploit the medicinal potential of EH, further studies on understanding the effective material basis and mechanisms are warranted.

Vision Transformers (ViT) is known for its scalability. In this work, we target to scale down a ViT to fit in an environment with dynamic-changing resource constraints. We observe that smaller ViTs are intrinsically the sub-networks of a larger ViT with different widths. Thus, we propose a general framework, named Scala, to enable a single network to represent multiple smaller ViTs with flexible inference capability, which aligns with the inherent design of ViT to vary from widths. Concretely, Scala activates several subnets during training, introduces Isolated Activation to disentangle the smallest sub-network from other subnets, and leverages Scale Coordination to ensure each sub-network receives simplified, steady, and accurate learning objectives. Comprehensive empirical validations on different tasks demonstrate that with only one-shot training, Scala learns slimmable representation without modifying the original ViT structure and matches the performance of Separate Training. Compared with the prior art, Scala achieves an average improvement of 1.6% on ImageNet-1K with fewer parameters.

This study investigated the potential of various detergents as carbon substrates for polyhydroxyalkanoate (PHA) production under nitrogen-limited conditions. We determined that Tween 80 promoted PHA accumulation by ZD-F1, a Bacillus sp. strain isolated from field sludge. When cultured with Tween 80, ZD-F1 achieved a peak dry cell weight (DCW) of 3.28 g/L, along with a PHA content of 18.96% and PHA yield of 0.48 g/L. Tween 80 at various concentrations resulted in a PHA theoretical yield of 0–10.6%. Experiments assessing the PHA species indicated that the PHA species under both Tween 80 and glucose was 3HB. Moreover, through chemical and biochemical stoichiometric calculations, we determined that the theoretical yields of 3HB from Tween 80 were 1.228 and 0.558 g/g, respectively, and those of 3HB from glucose were 0.716 and 0.318–0.477 g/g, respectively, across various catabolic pathways. To our knowledge, this is the first study to analyze PHA accumulation by a Bacillus sp. strain—isolated from wastewater treatment plant sludge—under various detergents as carbon sources.

Diffusion model based language-guided image editing has achieved great success recently. However, existing state-of-the-art diffusion models struggle with rendering correct text and text style during generation. To tackle this problem, we propose a universal self-supervised text editing diffusion model (DiffUTE), which aims to replace or modify words in the source image with another one while maintaining its realistic appearance. Specifically, we build our model on a diffusion model and carefully modify the network structure to enable the model for drawing multilingual characters with the help of glyph and position information. Moreover, we design a self-supervised learning framework to leverage large amounts of web data to improve the representation ability of the model. Experimental results show that our method achieves an impressive performance and enables controllable editing on in-the-wild images with high fidelity. Our code will be avaliable in \url{https://github.com/chenhaoxing/DiffUTE}.

With the popularity of mobile internet terminals, the speed of the network and With the popularization of mobile Internet terminals, the speed of network and the reduction of traffic tariff, people can watch videos through cell phones at any time. As short videos have the characteristics of short time and rich content, they can maximize people's need to watch videos in fragmented time. Therefore, short videos that combine filming techniques, music, stories and images. Short videos can meet users' content consumption needs in a more diversified way. Due to the widespread popularity of short videos, many short video platforms have been born. Many short video platforms have been created. The short video recommendation algorithm has become an important means of competition among platforms. How to ensure the accuracy of the recommendation algorithm and the real-time of the algorithm How to ensure the accuracy of the recommendation algorithm and the real-time of the algorithm have been the focus of research. In recent years, a great deal of progress has been me in the study of problems related to preference recommendation and recommendation. The most widely used method in industry today is the use of LR to learn click-through recommendation models. LR has the vantage of being simple and very easy to implement for massively real-time parallel processing, but linear models have a limited learning capability and do not capture the information carried by higher-order features , thus limiting the recommendation performance. Based on the above analysis, this paper proposes a FM-based short video preference recommendation model from the multi-topic nature of short videos. The main contributions of the model are: Topic-based segmentation of the original training set is performed using LDA, and each sub-training set generated by the segmentation is significantly smaller than the original training set, which reduces the computational complexity to a certain extent. Automated feature selection and linear transformation of features for training sets under different topics, reducing the dependence of manual feature engineering on time and labor in the baseline algorithm. By integrating the recommendation results of different topics, thus improving the recommendation accuracy. Experiments demonstrate that our model can effectively improve the recommendation performance.

Magnetic tunnel junction (MTJ) has been successfully used in spintronic devices, such as magnetoresistive random access memory, tunneling magnetoresistance (TMR) sensor, magnetic logic. TMR sensors use a special magnetic structure between the free layer and the pinned layer of an MTJ to realize a linear output. So far, five types of TMR linear sensing units (TMR-LSNs) have been developed based on MgO MTJs, which are artificial-indirect-double-exchange-coupling-, magnetic-field-biased-, in-plane-/perpendicular- and superparamagnetic- TMR-LSNs, respectively. In this paper, the five types of TMR-LSNs are combed and their magnetic sensing performance is systematically compared. First, the five types of TMR-LSNs have a linear resistance response with the external magnetic field with a changeable sensitivity, a linear field range and a low frequency noise level. Second, the five types of TMR-LSNs use different magnetic structures to realize the same aim that is to obtain the optimized performance parameters, which is of significance for the application of TMR sensors. Third, the five types of TMR-LSNs are suitable for different application scenarios due to their respective performance parameters. Therefore, we believe our summarized discussion in this paper would provide help for people to explore and find the related applications of TMR sensors based on the five types of TMR-LSNs.