Medical vision–language pretraining typically relies on static image–text pairs, overlooking temporal cues vital for understanding clinical progression. This limits model sensitivity to evolving semantics and reduces their effectiveness in real-world clinical reasoning. To address this challenge, we propose TAMM—a temporal alignment framework that leverages weak but semantically rich supervision from large language models (LLMs). Given temporally adjacent clinical reports, LLMs automatically generate (i) coarse-grained trend labels (e.g., improving or worsening), and (ii) fine-grained rationales explaining the supporting clinical evidence. These complementary signals inject temporal semantics without requiring manual annotation, and guide vision–language representation learning to capture trend-sensitive cross-modal alignment and rationale-grounded coherence. Experiments on multiple medical benchmarks demonstrate that TAMM improves retrieval and classification performance while yielding more interpretable, temporally consistent embeddings. Our results highlight the potential of leveraging LLM-derived supervision to equip vision–language models with temporal awareness critical for clinical applications.

Establishing local semantic correspondences between medical images and their corresponding reports is crucial for effective medical vision-language pre-training. However, existing methods encounter two major challenges: (1) lesion regions in radiological images are often small, blurry, or lack clear boundaries, complicating accurate localization; and (2) medical reports typically contain redundant or non-diagnostic words, hindering precise semantic alignment. To overcome these issues, we propose MedAligner, a specialized local alignment network for medical vision-language pre-training. MedAligner employs dual encoders to extract both global and local representations and uses global contrastive learning to maintain coarse semantic consistency. To enhance local alignment, we introduce a Word-Region Alignment, which generates a learnable word-pixel similarity matrix that is sparsified to identify salient lesion regions accurately. Additionally, our Diagnostic Term Filtering dynamically samples high-importance diagnostic terms from reports, aligning them with identified lesion areas via a local contrastive loss. Importantly, we adopt a progressive training strategy that gradually refines both the input text and semantic alignment. This is achieved by reconstructing concise diagnostic reports and progressively updating word-pixel similarity, generating increasingly accurate image-text pairs. Extensive experiments demonstrate that MedAligner significantly surpasses existing approaches on tasks such as phrase grounding, image-text retrieval, and zero-shot classification, setting new benchmarks in medical vision-language pre-training.

The increasing interest in learning from paired medical images and textual reports highlights the need for methods that can achieve multi-grained alignment between these two modalities. However, most existing approaches overlook finegrained semantic alignment, which can constrain the quality of the generated representations. To tackle this problem, we propose the Multi-Grained Vision-and-Language Alignment (MGVLA) model, which effectively leverages multi-grained correspondences between medical images and texts at different levels, including disease, instance, and token levels. For disease-level alignment, our approach adopts the concept of contrastive learning and uses medical terminologies detected from textual reports as soft labels to guide the alignment process. At the instance level, we propose a strategy for sampling hard negatives, where images and texts with the same disease type but differing in details such as disease locations and severity are considered as hard negatives. This strategy helps our approach to better distinguish between positive and negative image-text pairs, ultimately enhancing the quality of our learned representations. For token-level alignment, we employ a masking and recovery technique to achieve finegrained semantic alignment between patches and sub-words. This approach effectively aligns the different levels of granularity between the image and language modalities. To assess the efficacy of our MGVLA model, we conduct comprehensive experiments on the image-text retrieval and phrase grounding tasks.

The Corona Virus Disease 2019 (COVID-19) epidemic is a sudden public health crisis, known as an "International Emergency of Public Health Event". This study uses the bottom-up characteristics of multi-agents to construct multi-agent simulation models for COVID-19 prevention and control. The development trend of the epidemic situation under the condition that the government adopts different prevention and control measures is studied, and on this basis, the influence of temperature on the spread of the virus is discussed. The simulation results show that the multi-agent modeling method can effectively capture the emergence of complex systems. The evaluation of the effects of single measures and multiple interventions will help determine key prevention and control strategies and provide important experience and scientific basis for future epidemic prevention and control.

It is of great significance to predict soil organic carbon (SOC) stock of North China Plain which is one of the most important grain production bases with wheat-corn rotation as its main cropping system in China. In this study, a data simulation strategy is developed to predict SOC stock which is featured by the syntheses of two models, Estimation System for Agricultural Productivity (ESAP) and the spatial explicit SOC decomposition model. Model-based simulation indicates that the SOC stock in the agricultural soils of North China Plain would change in a nonlinear trend in the next two decades.

Detecting the situation of cropland in the semi-arid area, north of China is very important for agricultural management, land degradation and ecosystem assessment. In this study we tried to explore the potential and the methodology for the cropland identification using high temporal resolution SPOT-4 VEGETATION (VGT) imagery and some ancillary data. The imagery used in this study were 10-day composite Normalized Difference Vegetation Index (NDVI) over the year 2000 obtained by the Maximum Value Composite (MVC) technique, and the ancillary data included phonological calendar and agricultural knowledge. The discrete Fourier transform was applied to the NDVI data set on a per pixel basis for the whole study areas and then the additive and the first four harmonics (amplitude and phase) were classified using ISOLATE unsupervised classification algorithm. Analysis of the characteristics of NDVI temporal profiles reconstructed from the first four harmonics and the phonological calendar over the plant growing season allowed for the identification of distinct growth patterns between the different vegetation types. The accuracy of the result was evaluated with the agricultural census data and existing land-cover dataset derived from TM images. The result of this study shows that the methodology used in this study is, in general, feasible for cropland identification in semi-arid area of north China.