Some pathogens colonize plant leaves, but others invade the roots, including the vasculature, causing severe disease symptoms. Plant innate immunity has been extensively studied in leaf pathosystems; however, the precise regulation of immunity against vascular pathogens remains largely unexplored. We previously demonstrated that loss of function of the receptor kinase FERONIA (FER) increases plant resistance to the typical vascular bacterial pathogen Ralstonia solanacearum. Here, we show that upon infection with R. solanacearum, root xylem cell walls in Arabidopsis thaliana become highly lignified. FER is specifically upregulated in the root xylem in response to R. solanacearum infection, and inhibits lignin biosynthesis and resistance to this pathogen. We determined that FER interacts with and phosphorylates the transcription factor RESPONSIVE TO DESICCATION 26 (RD26), leading to its degradation. Overexpression and knockout of RD26 demonstrated that it positively regulates plant resistance to R. solanacearum by directly activating the expression of lignin-related genes. Tissue-specific expression of RD26 in the root xylem confirmed its role in vascular immunity. We confirmed that the FER–RD26 module regulates lignin biosynthesis and resistance against R. solanacearum in tomato (Solanum lycopersicum). Taken together, our findings unveil that the FER–RD26 cascade governs plant immunity against R. solanacearum in vascular tissues by regulating lignin deposition. This cascade may represent a key defense mechanism against vascular pathogens in plants.

With the rapid growth of urban vehicles, it is necessary to use advanced traffic control technologies for traffic flow control and traffic diversion, and allocate road resources reasonably to ensure smooth urban traffic and maximize the transportation efficiency of the transportation network. Considering the spatial characteristics of urban road networks and the temporal information of traffic speeds, this paper proposes a spatiotemporal dynamic deep network called GCN-LSTM, which is based on Graph Convolutional Network (GCN) and Long Short Term Memory (LSTM) neural network. The network utilizes a unique network structure to address the spatiotemporal dynamics, periodicity, and uncertainty of traffic flow. The proposed algorithm utilizes GCN to process the spatial correlation characteristics of traffic flow, and captures the temporal dynamic characteristics of traffic flow using LSTM and attention mechanism. Attention vectors are used as competitive selection to establish the correlation between the state of the next prediction step and the historical state. When predicting traffic fluctuations, this paper considers the spatiotemporal correlation in the road network and extracts spatiotemporal correlation features from the fluctuations. Finally, the proposed algorithm uses LSTM to fit historical features and future traffic flow fluctuation momentum. Simulation experiments have demonstrated the effectiveness and superiority of the proposed model.

As the challenges of globalization and climate change intensify, the importance of urban resilience in city planning is becoming increasingly evident. To adapt to this trend, innovations and improvements are essential in traditional urban land-use patterns to better fulfill the requirements of resilient urban development. In this context, this study constructs an urban resilience evaluation index system from four perspectives: social resilience, engineering resilience, ecological resilience, and security resilience to evaluate the urban resilience of Changsha City. A thorough assessment of the resilience mechanisms in Changsha’s urban layout was conducted, employing the SD-FLUS model. A resilient urban scenario is also established to restrict the conversion of high-resilience land into other land types and to predict urban land-use structures under a resilience-oriented directive. The findings indicate that areas with high ecological and safety resilience in Changsha are primarily located in the western Weishan mountain system, along with eastern mountain systems like Jiuling, Lianyun, and Mufu, forming the “green veins”. The central areas are characterized by “blue veins”, mainly represented by rivers such as the Xiangjiang, Weishui, Longwanggang, Jinjiang, Liuyang, and Laodao. Within the central urban area, high-resilience regions are primarily distributed along a framework consisting of “one ring (the city’s three-ring line), two mains (Xiangjiang and Liuyang rivers), one heart (urban green core), and six wedges”, specifying various green corridors. Under the resilience-oriented scenario, the model predicts that by 2025, the total built-up area in Changsha will be 1416.79 km². Areas with high social and engineering resilience are mainly concentrated in the central urban areas of Changsha, as well as Ningxiang and Liuyang, aligning closely with the objectives of Changsha’s latest round of national spatial planning. The built-up area layout should complement Changsha’s topography and water systems, expanding in a wedge-like manner. Overall, Changsha’s planning has successfully integrated social, engineering, ecological, and safety resilience, enhancing its adaptability and long-term sustainability. This research proposes a land-use simulation method guided by the concept of urban resilience, providing valuable insights for resilience-oriented city planning in Changsha and other cities facing similar challenges.

Urban parks play an increasingly important role in supporting social interaction, ecological services, and everyday well-being in rapidly urbanizing cities, yet prevailing planning practices still rely on equal-provision logics and linear modeling frameworks, implicitly assuming that park vitality increases proportionally with facilities and surrounding services. Such assumptions overlook the possibility that park vitality responds to built-environment factors in nonlinear, threshold-based, and configuration-dependent ways. This study develops an interpretable machine learning approach to identify the nonlinear effects and structural configurations that drive urban park vitality in Changsha, China. We integrate Baidu Huiyan population heat data with AOI-defined park boundaries and multi-source POI indicators to characterize internal facilities and surrounding built-environments for 147 parks in the city’s main urban area. An XGBoost model is trained to predict park vitality, and SHAP values, partial dependence analysis, and bivariate interaction plots are employed to examine variable importance, threshold behaviors, and synergistic or substitutive relationships among key factors. The results show that sports and leisure facilities are the most influential driver of vitality, followed by shopping services and government service facilities. Their impacts are strongly nonlinear: sports and leisure facilities and public amenities display clear saturation thresholds, while high-density shopping services generate substantial gains in vitality only beyond specific concentration levels. Interaction effects further indicate that park vitality emerges from particular configurations of internal facilities and surrounding residential and service environments, rather than from the additive accumulation of isolated factors. These findings demonstrate the value of interpretable machine learning for shifting urban park planning from equal-provision paradigms toward structurally informed configuration strategies and more efficient public space governance.

Video Question Answering (VideoQA) requires a deep understanding of dynamic video content, integrating spatial reasoning, temporal dependencies, and language comprehension. Existing methods often struggle with long or semantically complex videos due to the lack of question-guided keyframe weight adjustment and the absence of question-aligned cross-modal description generation. To address these challenges, we propose ETR (Event-centric Temporal Reasoning), an adaptive framework for VideoQA. ETR introduces three key mechanisms: (i) a hierarchical weight adjustment selector to identify questions requiring event-centric temporal reasoning; (ii) a T-Route that segments videos into semantically coherent events and dynamically adjusts keyframe weights with question intent; and (iii) a question-conditioned prompting strategy that focuses on key objects to generate textual prompts aligned with a question’s semantics. This hierarchical and adaptive design effectively balances visual and textual information, enhances temporal reasoning, and improves object-centric alignment. Experiments on two datasets demonstrate that ETR achieves competitive performance in fine question-aware VideoQA.

The urban construction land use scale is an important aspect of urban research, and simulating the scale and spatial layout of urban development land use is of great significance for sustainable development of urban areas. Currently, traditional methods for predicting urban land use scale lack dynamic research on the relationship between factors that affect urban land use scale, resulting in predictions that deviate from actual situations. In this paper, we combine the advantages of the system dynamics model and the FLUS model to improve traditional methods for predicting urban land use scale. Using Changsha City as a research area, we conducted a prediction of urban land use scale in 2020, and adjusted the model based on multiple years of data. The simulation results were accurate to 0.94 in terms of correlation coefficient (kappa = 0.90). We then applied this model and method to simulate urban land use scale in 2025 in Changsha City, and discussed the feasibility of using this simulation to provide reference for land development and land resource protection.