LLM-powered multi-agent systems (MAS) have demonstrated strong performance on complex tasks.However, most existing approaches still rely on hand-crafted communication protocols or automatically designed communication topologies, which generalize poorly across tasks.We introduce NeuralFSM, a state-driven framework that formulates multi-agent problem solving as a finite-state execution process.NeuralFSM learns both the state transition distribution and inter-agent communication weights from interaction traces using a Temporal Coordination Controller.Rather than prioritizing explicit structure generation, the proposed framework uses task context to modulate transition and routing decisions, enabling flexible coordination without manual protocol design.To improve robustness against noisy or adversarial agents, we incorporate graph regularization during training and apply trust-aware message attenuation at runtime.Experiments on diverse benchmarks show that NeuralFSM consistently outperforms previous baselines by an average margin of 6.74% ∼ 19.39%, while substantially reducing token consumption.Moreover, NeuralFSM exhibits strong inherent robustness, which is further enhanced by the protection layer, resulting only in a 1.82% performance drop under attack.The code is available at https: //github.com/DisseverYOLO/NeuralFSM.

Nuclear receptors (NRs) are ligand-activated transcription factors, which constitute one of the most important targets for drug discovery. Current computational strategies mainly focus on a single target, and the transfer of learned knowledge among NRs was not considered yet. Herein we proposed a novel computational framework named NR-Profiler for prediction of potential NR modulators with high affinity and specificity. First, we built a comprehensive NR data set including 42 684 interactions to connect 42 NRs and 31 033 compounds. Then, we used multi-task deep neural network and multi-task graph convolutional neural network architectures to construct multi-task multi-classification models. To improve the predictive capability and robustness, we built a consensus model with an area under the receiver operating characteristic curve (AUC) = 0.883. Compared with conventional machine learning and structure-based approaches, the consensus model showed better performance in external validation. Using this consensus model, we demonstrated the practical value of NR-Profiler in virtual screening for NRs. In addition, we designed a selectivity score to quantitatively measure the specificity of NR modulators. Finally, we developed a freely available standalone software for users to make profiling predictions for their compounds of interest. In summary, our NR-Profiler provides a useful tool for NR-profiling prediction and is expected to facilitate NR-based drug discovery.

Drug discovery and development is a time-consuming and costly process. Therefore, drug repositioning has become an effective approach to address the issues by identifying new therapeutic or pharmacological actions for existing drugs. The drug's anatomical therapeutic chemical (ATC) code is a hierarchical classification system categorized as five levels according to the organs or systems that drugs act and the pharmacology, therapeutic and chemical properties of drugs. The 2nd-, 3rd- and 4th-level ATC codes reserved the therapeutic and pharmacological information of drugs. With the hypothesis that drugs with similar structures or targets would possess similar ATC codes, we exploited a network-based approach to predict the 2nd-, 3rd- and 4th-level ATC codes by constructing substructure drug-ATC (SD-ATC), target drug-ATC (TD-ATC) and Substructure&Target drug-ATC (STD-ATC) networks. After 10-fold cross validation and two external validations, the STD-ATC models outperformed the SD-ATC and TD-ATC ones. Furthermore, with KR as fingerprint, the STD-ATC model was identified as the optimal model with AUC values at 0.899 ± 0.015, 0.916 and 0.893 for 10-fold cross validation, external validation set 1 and external validation set 2, respectively. To illustrate the predictive capability of the STD-ATC model with KR fingerprint, as a case study, we predicted 25 FDA-approved drugs (22 drugs were actually purchased) to have potential activities on heart failure using that model. Experiments in vitro confirmed that 8 of the 22 old drugs have shown mild to potent cardioprotective activities on both hypoxia model and oxygen-glucose deprivation model, which demonstrated that our STD-ATC prediction model would be an effective tool for drug repositioning.

Apache Hive has been widely used for big data processing over large scale clusters by many companies. It provides a declarative query language called HiveQL. The efficiency of filtering out query-irrelevant data from HDFS closely affects the performance of query processing. This is especially true for multi-dimensional, high-selective, and few columns involving queries, which provides sufficient information to reduce the amount of bytes read. Indexing (Compact Index, Aggregate Index, Bitmap Index, DGFIndex, and the index in ORC file) and columnar storage (RCFile, ORC file, and Parquet) are powerful techniques to achieve this. However, it is not trivial to choosing a suitable index and columnar storage based on data and query features. In this paper, we compare the data filtering performance of the above indexes with different columnar storage formats by conducting comprehensive experiments using uniform and skew TPC-H data sets and various multi-dimensional queries, and suggest the best practices of improving multi-dimensional queries in Hive under different conditions.

Current QoS-aware automatic service composition queries over a network of web services are often one-time in nature. After a network of web services is built, such queries are issued once, and answers are found from the scratch. The underlying assumption is that the participating web services are rather static, which means that their functional and non-functional parameters seldom change. However, such an assumption is often baseless. New services come and go, service APIs change gradually, and QoS values fluctuate in practice. Therefore, a support for efficiently adapting service composition is desired. In this paper, we propose an event driven continuous query algorithm to intelligently cope with different types of dynamic services, and thus enable evolution of service composition. Evaluation using both real QoS data and synthetic web service data shows its superior performance, compared with the state-of-the-art solution which won the performance championship of Web Service Challenge in 2009 and 2010. Moreover, we generalize a new graph problem: dynamic single-source optimal-Directed Acyclic Graphs (DAG) problem based on the above work. It is argued that API recommendation for framework evolution, as another typical application, could also be modeled and addressed efficiently using the proposed new graph approach.

This paper introduces a robust method for computing the optimal route with hierarchy. We convert a planar road network into its Voronoi-based counterpart with multiple levels of detail (LoDs), which is subsequently assigned travel times that are estimated for different times of day using taxicab trajectory data. On the basis of this network structure, we model the path-finding process in travel, as the optimal route with hierarchy is computed in a ‘coarse-to-fine’ manner. In other words, the route is iteratively constructed from roads in a low LoD network to roads in a high LoD network. To confirm the efficiency and effectiveness of our method, comparative experiments were conducted using randomly selected pairs of origins/destinations in Wuhan, China. The results indicate that our travel lengths are on average 12% longer than those computed by the Dijkstra algorithm and 15% shorter than those computed by the hierarchical algorithm (in ArcGIS). Our travel times are on average 29% longer than those computed by the Dijkstra algorithm and 31% shorter than those computed by the hierarchical algorithm (in ArcGIS). Hence, we argue that our method is situated in terms of performance between the Dijkstra algorithm and the hierarchical algorithm (in ArcGIS). Moreover, road usage patterns confirm that our algorithm is cognitively equivalent to the hierarchical algorithm (in ArcGIS) by favoring high-class roads and outperforms the Dijkstra algorithm by avoiding choosing low-class roads. Computationally, our method outperforms the Dijkstra algorithm but is on the same level as the hierarchical algorithm (in ArcGIS) in terms of efficiency. Therefore, it has the potential to be used in real-time routing applications or services.

In this paper we propose a two-stage algorithm for robust K-subspaces recovery. In the first stage, a large number of local candidate subspaces are generated by probabilistic farthest insertion, and then the initial near-optimal K-subspaces are solved by combinatorial selection with randomized greedy method. In the second stage, the K-subspaces are further refined by assigning each data vector to the closest subspace and taking proper gradient steps along the geodesic of Grassmannian to update the subspace. Numerical experiments show that our two-stage algorithm can robustly recover K-subspaces even with large fraction of outliers. Experiments on the well-known Hopkins 155 dataset of motion segmentation show that the proposed approach significantly outperforms the baseline LBF algorithm, and can compete with state of the art subspace clustering algorithms with less running time.

Hive is the most mature and prevalent data warehouse tool providing SQL-like interface in the Hadoop ecosystem. It is successfully used in many Internet companies and shows its value for big data processing in traditional industries. However, enterprise big data processing systems as in Smart Grid applications usually require complicated business logics and involve many data manipulation operations like updates and deletes. Hive cannot offer sufficient support for these while preserving high query performance. Hive using the Hadoop Distributed File System (HDFS) for storage cannot implement data manipulation efficiently and Hive on HBase suffers from poor query performance even though it can support faster data manipulation. There is a project based on Hive issue Hive-5317 to support update operations, but it has not been finished in Hive's latest version. Since this ACID compliant extension adopts same data storage format on HDFS, the update performance problem is not solved. In this paper, we propose a hybrid storage model called DualTable, which combines the efficient streaming reads of HDFS and the random write capability of HBase. Hive on DualTable provides better data manipulation support and preserves query performance at the same time. Experiments on a TPC-H data set and on a real smart grid data set show that Hive on DualTable is up to 10 times faster than Hive when executing update and delete operations.

In this letter, an effective enhancement method for remote sensing images is introduced to improve the global contrast and the local details. The proposed method constitutes an empirical approach by using the regularized-histogram equalization (HE) and the discrete cosine transform (DCT) to improve the image quality. First, a new global contrast enhancement method by regularizing the input histogram is introduced. More specifically, this technique uses the sigmoid function and the histogram to generate a distribution function for the input image. The distribution function is then used to produce a new image with improved global contrast by adopting the standard lookup table-based HE technique. Second, the DCT coefficients of the previous contrast improved image are automatically adjusted to further enhance the local details of the image. Compared with conventional methods, the proposed method can generate enhanced remote sensing images with higher contrast and richer details without introducing saturation artifacts.