Subseasonal-to-seasonal (S2S) forecasting, which predicts climate conditions from several weeks to months in advance, represents a critical frontier for agricultural planning, energy management, and disaster preparedness. However, it remains one of the most challenging problems in atmospheric science, due to the chaotic dynamics of atmospheric systems and complex interactions across multiple scales. Current approaches often fail to explicitly model underlying physical processes and teleconnections that are crucial at S2S timescales. We introduce TelePiT, a novel deep learning architecture that enhances global S2S forecasting through integrated multi-scale physics and teleconnection awareness. Our approach consists of three key components: (1) Spherical Harmonic Embedding, which accurately encodes global atmospheric variables onto spherical geometry; (2) Multi-Scale Physics-Informed Neural ODE, which explicitly captures atmospheric physical processes across multiple learnable frequency bands; (3) Teleconnection-Aware Transformer, which models critical global climate interactions through explicitly modeling teleconnection patterns into the self-attention. Extensive experiments demonstrate that TelePiT significantly outperforms state-of-the-art data-driven baselines and operational numerical weather prediction systems across all forecast horizons, marking a significant advance toward reliable S2S forecasting.

Nuclear radiation, which refers to the energy emitted from atomic nuclei during decay, poses significant risks to human health and environmental safety. Recently, advancements in monitoring technology have facilitated the effective recording of nuclear radiation levels and related factors, such as weather conditions. The abundance of monitoring data enables the development of accurate and reliable nuclear radiation forecasting models, which play a crucial role in informing decision-making for individuals and governments. However, this task is challenging due to the imbalanced distribution of monitoring stations over a wide spatial range and the non-stationary radiation variation patterns. In this study, we introduce NRFormer, a novel framework tailored for the nationwide prediction of nuclear radiation variations. By integrating a non-stationary temporal attention module, an imbalance-aware spatial attention module, and a radiation propagation prompting module, NRFormer collectively captures complex spatio-temporal dynamics of nuclear radiation. Extensive experiments on two real-world datasets demonstrate the superiority of our proposed framework against 11 baselines. NRFormer has been deployed online to provide 1-24-day nuclear radiation forecasts, empowering individuals and governments with timely, data-driven decisions for emergency response and public safety. Our framework is designed for general applicability and can be readily adapted for deployment in other regions. The deployed system is available at https://NRFormer.github.io and the dataset and code of the predictive model are available at https://github.com/usail-hkust/NRFormer.

Spatio-temporal forecasting is a critical component of various smart city applications, such as transportation optimization, energy management, and socio-economic analysis. Recently, several automated spatio-temporal forecasting methods have been proposed to automatically search the optimal neural network architecture for capturing complex spatio-temporal dependencies. However, the existing automated approaches suffer from expensive neural architecture search overhead, which hinders their practical use and the further exploration of diverse spatio-temporal operators in a finer granularity. In this paper, we propose AutoSTF, a decoupled automatic neural architecture search framework for cost-effective automated spatio-temporal forecasting. From the efficiency perspective, we first decouple the mixed search space into temporal space and spatial space and respectively devise representation compression and parameter-sharing schemes to mitigate the parameter explosion. The decoupled spatio-temporal search not only expedites the model optimization process but also leaves new room for more effective spatio-temporal dependency modeling. From the effectiveness perspective, we propose a multi-patch transfer module to jointly capture multi-granularity temporal dependencies and extend the spatial search space to enable finer-grained layer-wise spatial dependency search. Extensive experiments on eight datasets demonstrate the superiority of AutoSTF in terms of both accuracy and efficiency. Specifically, our proposed method achieves up to 13.48x speed-up compared to state-of-the-art automatic spatio-temporal forecasting methods while maintaining the best forecasting accuracy. The source code and data are available at https://github.com/usail-hkust/AutoSTF.

Foundation models (FMs), such as GPT-4 and AlphaFold, are reshaping the landscape of scientific research. Beyond accelerating tasks such as hypothesis generation, experimental design, and result interpretation, they prompt a more fundamental question: Are FMs merely enhancing existing scientific methodologies, or are they redefining the way science is conducted? In this paper, we argue that FMs are catalyzing a transition toward a new scientific paradigm. We introduce a three-stage framework to describe this evolution: (1) Meta-Scientific Integration, where FMs enhance workflows within traditional paradigms; (2) Hybrid Human-AI Co-Creation, where FMs become active collaborators in problem formulation, reasoning, and discovery; and (3) Autonomous Scientific Discovery, where FMs operate as independent agents capable of generating new scientific knowledge with minimal human intervention. Through this lens, we review current applications and emerging capabilities of FMs across existing scientific paradigms. We further identify risks and future directions for FM-enabled scientific discovery. This position paper aims to support the scientific community in understanding the transformative role of FMs and to foster reflection on the future of scientific discovery.

The Collective Perception Service (CPS) enables vehicles and infrastructure to cooperatively share local sensor measurements to extend their perception range for a more comprehensive understanding of their environment. Although CPS transmissions are useful, their frequency and redundancy can saturate the vehicular communication channel hindering transmissions of other critical messages. To mitigate this, ETSI has specified generation rules for invoking Collective Perception Messages (CPMs), which are mainly based on objects’ dynamics and frequency of reporting but do not account for the perceptual accuracy of measurements. This can lead to ETSI generation rules potentially being invoked more frequently for inaccurately sensed objects, unnecessarily consuming network resources. Furthermore, inaccurate sensor data is more likely to be matched to the incorrect object track, yielding information that can mislead decision-making. The CAR-2-CAR Communication Consortium (C2C-CC) has proposed an object quality pre-filter to remove poor-quality measurements before triggering the collective perception generation rules, but its complexity has hindered deployment. At the request of ETSI and to hasten deployment, this paper explores the feasibility of proposing a ’simple’ pre-filter that will attempt to discard measurements that would be falsely associated based on a position error measure-based threshold. We find that implementing such a filter is complex and highly dependent on the environmental topology, road user type, geo-spatial density and temporal factors. Specifically, we investigate false associations under different sensor error ranges for diverse road users across 4 urban intersections, detailing their causation. We discuss the challenges in setting a distance measure-based threshold and evaluate its implications as a pre-filter to reduce the number of false and weakly associated objects in a CPM.

Recently, graph neural architecture search (GNAS) frameworks have been successfully used to automatically design the optimal neural architectures for many problems such as node classification and graph classification. In the existing GNAS frameworks, the designed graph neural network (GNN) architectures learn the representation of homogenous graphs with one type of relationship connecting two nodes. However, multi-view graphs, where each view represents a type of relationship among nodes, are ubiquitous in the real world. The traditional GNAS frameworks learn the graph representation without considering the interactions between nodes and multiple relationships, so they fail to solve multi-view graph-based problems, such as multi-view graphs modelling the biomedical entity and relation extraction tasks. In this paper, we propose MVGNAS, a multi-view graph neural network automatic modelling framework for biomedical entity and relation extraction, to resolve this challenge. In MVGNAS, we propose an automatic multi-view representation learning to learn low-dimensional representations of nodes that capture multiple relationships in a multi-view graph, representing the first research work in literature to solve the problem of multi-view graph representation learning architecture search for biomedical entity and relation extraction tasks. The experimental results demonstrate that MVGNAS can achieve the best performance in biomedical entity and relation extraction tasks against the state-of-the-art baseline methods.

It is significant to comprehend the relationship between metabolic pathway and molecular pathway for synthesizing new molecules, for instance optimizing drug metabolization. In bioinformatics fields, multi-label prediction of metabolic pathways is a typical manner to understand this relationship. Graph neural networks (GNNs) have become an effective method to extract molecular structure's features for multi-label prediction of metabolic pathways. Though GNNs can effectively capture structural features from molecular structure graphs, building a well-performed GNN model for a given molecular structure data set requires the manual design of the GNN architecture and fine-tuning of the hyperparameters, which are time-consuming and rely on expert experience. To address the above challenge, we design an end-to-end automatic molecular structure representation learning framework named AutoMSR that can design the optimal GNN model based on a given molecular structure data set without manual intervention. We propose a multi-seed age evolution (MSAE) search algorithm to identify the optimal GNN architecture from the GNN architecture subspace. For a given molecular structure data set, AutoMSR first uses MSAE to search the GNN architecture, and then it adopts a tree-structured parzen estimator to obtain the best hyperparameters in the hyperparameters subspace. Finally, AutoMSR automatically constructs the optimal GNN model based on the best GNN architecture and hyperparameters to extract the molecular structure features for multi-label metabolic pathway prediction. We test the performance of AutoMSR on the real data set KEGG. The experiment results show that AutoMSR outperforms baseline methods on different multi-label classification evaluation metrics.

In recent years, cancer patients survival prediction holds important significance for worldwide health problems, and has gained many researchers attention in medical information communities. Cancer patients survival prediction can be seen the classification work which is a meaningful and challenging task. Nevertheless, research in this field is still limited. In this work, we design a novel Multimodal Graph Neural Network (MGNN)framework for predicting cancer survival, which explores the features of real-world multimodal data such as gene expression, copy number alteration and clinical data in a unified framework. Specifically, we first construct the bipartite graphs between patients and multimodal data to explore the inherent relation. Subsequently, the embedding of each patient on different bipartite graphs is obtained with graph neural network. Finally, a multimodal fusion neural layer is proposed to fuse the medical features from different modality data. Comprehensive experiments have been conducted on real-world datasets, which demonstrate the superiority of our modal with significant improvements against state-of-the-arts. Furthermore, the proposed MGNN is validated to be more robust on other four cancer datasets.

Graph neural networks (GNNs) have received much attention as GNNs have recently been successfully applied on non-euclidean data. However, artificially designed graph neural networks often fail to get satisfactory model performance for a given graph data. Graph neural architecture search effectively constructs the GNNs that achieve the expected model performance with the rise of automatic machine learning. The challenge is efficiently and automatically getting the optimal GNN architecture in a vast search space. Existing search methods serially evaluate the GNN architectures, severely limiting system efficiency. To solve these problems, we develop an Auto matic G raph N eural A rchitecture S earch framework (Auto-GNAS) with parallel estimation to implement an automatic graph neural search process that requires almost no manual intervention. In Auto-GNAS, we design the search algorithm with multiple genetic searchers. Each searcher can simultaneously use evaluation feedback information, information entropy, and search results from other searchers based on sharing mechanism to improve the search efficiency. As far as we know, this is the first work using parallel computing to improve the system efficiency of graph neural architecture search. According to the experiment on the real datasets, Auto-GNAS obtain competitive model performance and better search efficiency than other search algorithms. Since the parallel estimation ability of Auto-GNAS is independent of search algorithms, we expand different search algorithms based on Auto-GNAS for scalability experiments. The results show that Auto-GNAS with varying search algorithms can achieve nearly linear acceleration with the increase of computing resources.

Understanding the relationship between molecular structure and metabolic pathway classes is significant for optimizing drug metabolization. In bioinformatics, graph neural networks (GNNs) can effectively capture structural and semantic features for molecular representation. Graph neural networks have become an essential method to encode molecular structures for multi-label prediction of metabolic pathways. However, building a GNN model for a given molecular structure dataset requires the manual design of GNN structure and fine-tuning of the hyperparameters for the GNN model, which is time-consuming and relies on expert experience. In this paper, we design an automatic end-to-end molecular structure representation learning framework named Auto-MSR that can design a GNN model for molecular structure encoding with little manual intervention. For a given compound molecule structure dataset, Auto-MSR first uses an efficient pa rallel GNN structure search algorithm to identify the optimal GNN structure from the GNN structure subspace. Then, it adopts a tree-structured parzen estimator approach to obtain the best hyperparameters of the GNN model in the hyperparameters subspace. We test AutoMSR on the dataset KEGG based on the multi-label metabolic pathway prediction task. The comparing results show that AutoMSR outperforms state-of-art manual graph neural networks on performance.

The interaction of multiple drugs could lead to serious events, which causes injuries and huge medical costs. Accurate prediction of drug-drug interaction (DDI) events can help clinicians make effective decisions and establish appropriate therapy programs. Recently, many AI-based techniques have been proposed for predicting DDI associated events. However, most existing methods pay less attention to the potential correlations between DDI events and other multimodal data such as targets and enzymes. To address this problem, we propose a Multimodal Deep Neural Network (MDNN) for DDI events prediction. In MDNN, we design a two-pathway framework including drug knowledge graph (DKG) based pathway and heterogeneous feature (HF) based pathway to obtain drug multimodal representations. Finally, a multimodal fusion neural layer is designed to explore the complementary among the drug multimodal representations. We conduct extensive experiments on real-world dataset. The results show that MDNN can accurately predict DDI events and outperform the state-of-the-art models.

Graph neural architecture search has received a lot of attention as Graph Neural Networks (GNNs) has been successfully applied on the non-Euclidean data recently. However, exploring all possible GNNs architectures in the huge search space is too time-consuming or impossible for big graph data. In this paper, we propose a parallel graph architecture search (GraphPAS) framework for graph neural networks. In GraphPAS, we explore the search space in parallel by designing a sharing-based evolution learning, which can improve the search efficiency without losing the accuracy. Additionally, architecture information entropy is adopted dynamically for mutation selection probability, which can reduce space exploration. The experimental result shows that GraphPAS outperforms state-of-art models with efficiency and accuracy simultaneously.

Predicting the survival of cancer patients holds significant meaning for public health, and has attracted increasing attention in medical information communities. In this study, we propose a novel framework for cancer survival prediction named Multimodal Graph Neural Network (MGNN), which explores the features of real-world multimodual data such as gene expression, copy number alteration and clinical data in a unified framework. In order to explore the inherent relation, we first construct the bipartite graphs between patients and multimodal data. Subsequently, graph neural network is adopted to obtain the embedding of each patient on different bipartite graphs. Finally, a multimodal fusion neural layer is designed to fuse the features from different modal data. The output of our method is the classification of short term survival or long term survival for each patient. Experimental results on one breast cancer dataset demonstrate that MGNN outperforms all baselines. Furthermore, we test the trained model on lung cancer dataset, and the experimental results verify the strong robust by comparing with state-of-the-art methods.