Explore the vast range of titles available.

In recent years, with the development of the Internet, the attribution classification of APT malware remains an important issue in society. Existing methods have yet to consider the DLL link library and hidden file address during the execution process, and there are shortcomings in capturing the local and global correlation of event behaviors. Compared to the structural features of binary code, opcode features reflect the runtime instructions and do not consider the issue of multiple reuse of local operation behaviors within the same APT organization. Obfuscation techniques more easily influence attribution classification based on single features. To address the above issues, (1) an event behavior graph based on API instructions and related operations is constructed to capture the execution traces on the host using the GNNs model. (2) ImageCNTM captures the local spatial correlation and continuous long-term dependency of opcode images. (3) The word frequency and behavior features are concatenated and fused, proposing a multi-feature, multi-input deep learning model. We collected a publicly available dataset of APT malware to evaluate our method. The attribution classification results of the model based on a single feature reached 89.24% and 91.91%. Finally, compared to single-feature classifiers, the multi-feature fusion model achieves better classification performance.

Contemporary large-scale and systematic agricultural operations demand the collaborative efforts of multiple agricultural machines with distinct functionalities. However, the failure of a single agricultural machine during collaborative operations jeopardizes the entire undertaking. To address this challenge, this paper proposes a multi-machine collaborative dynamic job allocation method based on the improved ant colony algorithm. Initially, the improved ant colony algorithm is employed to determine the optimal solution for harvester scheduling. This solution is then fed into the data conversion algorithm to acquire the necessary unloading point information for transport vehicle scheduling. Subsequently, the improved ant colony algorithm is once again utilized to optimize the transport vehicle scheduling. In cases of agricultural machinery failure or changes in the operating environment, two distinct methods are employed based on the situation. The first involves double-layer rescheduling of both harvester and transport vehicles, while the second employs single-layer rescheduling exclusively for the transport vehicles, yielding the respective rescheduling results. The outcomes demonstrate that the proposed solution method effectively identifies the current optimal scheduling plan for both the harvester and transport vehicle in the event of malfunctions. Moreover, under the premise that the unproductive waiting time of the harvester is reduced to zero, and the number of transport vehicles is minimized, it achieves the minimization of operating time cost and transportation cost. This method exhibits significant potential for seamless integration into the practical application of unmanned farms, providing a foundation for addressing scheduling and management challenges in multi-agricultural machinery collaborative operations within complex farmland operating environments.

Transforming the task of information extraction into a machine reading comprehension (MRC) framework has shown promising results. The MRC model takes the context and query as the inputs to the encoder, and the decoder extracts one or more text spans as answers (entities and relationships) from the text. Existing approaches typically use multi-layer encoders, such as Transformers, to generate hidden features of the source sequence. However, increasing the number of encoder layers can lead to the granularity of the representation becoming coarser and the hidden features of different words becoming more similar, potentially leading to the model’s misjudgment. To address this issue, a new method called the multi-granularity attention multi-scale self-learning network (MAML-NET) is proposed, which enhances the model’s understanding ability by utilizing different granularity representations of the source sequence. Additionally, MAML-NET can independently learn task-related information from both global and local dimensions based on the learned multi-granularity features through the proposed multi-scale self-learning attention mechanism. The experimental results on two information extraction tasks, named entity recognition and entity relationship extraction, demonstrated that the method was superior to the method based on machine reading comprehension and achieved the best performance on the five benchmark tests.

Natural killer/T-cell lymphoma (NKTCL) is a malignancy with a poor prognosis. The Smoothened (Smo) protein is implicated in NKTCL growth. This study employed lentiviral vector-mediated Smo RNA interference (LV- Smo -RNAi) to silence the Smo gene in the human NKTCL cell line SNT8. Fluorescence microscopy, qRT-PCR, and Western blot verified the reduction of Smo mRNA and protein levels. The CCK-8 assay showed that Smo silencing inhibited cell proliferation. Flow cytometry with annexin V-PE/7-AAD double staining indicated an increased apoptosis rate. Moreover, the expression of GLI family zinc finger 1 (Gli1) and programmed death-ligand 1 (PD-L1) was downregulated. In vivo, xenotransplantation experiments demonstrated that Smo silencing led to slower tumor growth with reduced tumor volume and weight. Overall, Smo gene silencing holds great potential as a novel molecular-targeted therapy approach for NKTCL by effectively suppressing cell proliferation and promoting apoptosis. Highlights 1. This study is the first to report the anti-tumor effects of Smo gene silencing in the human NKTCL cell line SNT8 2. This research finds that Smo gene silencing significantly suppresses cell proliferation. 3. This research observes that Smo silencing significantly increases the apoptosis rate of NKTCL cells. 4. Smo silencing downregulates the expression of Gli1 and PD-L1, providing new insights into the underlying mechanisms. 5.This study provides a new molecular target for developing targeted therapies for NKTCL.

Topic models can extract consistent themes from large corpora for research purposes. In recent years, the combination of pretrained language models and neural topic models has gained attention among scholars. However, this approach has some drawbacks: in short texts, the quality of the topics obtained by the models is low and incoherent, which is caused by the reduced word frequency (insufficient word co-occurrence) in short texts compared to long texts. To address these issues, we propose a neural topic model based on SBERT and data augmentation. First, our proposed easy data augmentation (EDA) method with keyword combination helps overcome the sparsity problem in short texts. Then, the attention mechanism is used to focus on keywords related to the topic and reduce the impact of noise words. Next, the SBERT model is trained on a large and diverse dataset, which can generate high-quality semantic information vectors for short texts. Finally, we perform feature fusion on the augmented data that have been weighted by an attention mechanism with the high-quality semantic information obtained. Then, the fused features are input into a neural topic model to obtain high-quality topics. The experimental results on an English public dataset show that our model generates high-quality topics, with the average scores improving by 2.5% for topic coherence and 1.2% for topic diversity compared to the baseline model.

To enhance the synergistic effect of phosphorus (P) and nitrogen (N) on flame retardant property, four different phytic acid-based NP flame retardants (FR-PAN) were manufactured using phytic acid and urea with various molar ratios, ranging from 1:3 to 1:12. The FR-PAN water solution was used to impregnate poplar wood under vacuum condition, and the thermal degradation performance of the FR-PAN treated wood were investigated. Compared to untreated wood, the PAN-6 (molar ratio is 1:6) group showed a reduction of 57.1% in total heat release and 80.0% in total smoke release. In the combustion, due to the introduction of P and N, FR-PAN generates O=P/C-O/C-P/C-N bonds, forming highly graphitized char residues, which effectively isolate the entry of oxygen and heat and play a good protective role in the condensed phase. Morphological and chemical analysis of the residual char layer revealed that the introduction of P and N elements formed a more stable hybrid structure, significantly improving the thermal stability of the char layer. Among them, the PAN-6 group exhibited the highest char layer stability, indicating optimal synergistic effects of nitrogen and phosphorus under these conditions.

Brassinosteroids (BRs) play important roles in regulating gravitropism and vasculature development. Here, we report the effect of brassinosteroids on negative gravitropism and G-fiber cell wall development of the stem in woody angiosperms. We applied exogenous 24-epibrassinolide (BL) or its biosynthesis inhibitor brassinazole (BRZ) to slanted hybrid poplar trees (Populus deltoids × Populus nigra) and measured the morphology of gravitropic stems, anatomy and chemistry of secondary cell wall. We furthermore analyzed the expression levels of auxin transport and cellulose biosynthetic genes after 24-epibrassinolide (BL) or brassinazole (BRZ) application. The BL-treated seedlings showed no negative gravitropism bending, whereas application of BRZ dramatically enhanced negative gravitropic bending. BL treatment stimulated secondary xylem fiber elongation and G-fiber formation on the upper side of stems but delayed G-fiber maturation. BRZ inhibited xylem fiber elongation but induced the production of more mature G-fibers on the upper side of stems. Wood chemistry analyses and immunolocalization demonstrated that BL and BRZ treatments increased the cellulose content and modified the deposition of cell wall carbohydrates including arabinose, galactose and rhamnose in the secondary xylem. The expression of cellulose biosynthetic genes, especially those related to cellulose microfibril deposition (PtFLA12 and PtCOBL4) was significantly upregulated in BL-and BRZ-treated TW stems compared with control stems. The significant differences of G-fibers development and negative gravitropism bending between 24-epibrassinolide (BL) and brassinazole (BRZ) application suggest that brassinosteroids are important for secondary xylem development during tension wood formation. Our findings provide potential insights into the mechanism by which BRs regulate G-fiber cell wall development to accomplish negative gravitropism in TW formation.

In the field of Chinese-named entity recognition, recent research has sparked new interest by combining lexical features with character-based methods. Although this vocabulary enhancement method provides a new perspective, it faces two main challenges: firstly, using character-by-character matching can easily lead to conflicts during the vocabulary matching process. Although existing solutions attempt to alleviate this problem by obtaining semantic information about words, they still lack sufficient temporal sequential or global information acquisition; secondly, due to the limitations of dictionaries, there may be words in a sentence that do not match the dictionary. In this situation, existing vocabulary enhancement methods cannot effectively play a role. To address these issues, this paper proposes a method based on lexical information and spatial features. This method carefully considers the neighborhood and overlap relationships of characters in vocabulary and establishes global bidirectional semantic and temporal sequential information to effectively address the impact of conflicting vocabulary and character fusion on entity segmentation. Secondly, the attention score matrix extracted by the point-by-point convolutional network captures the local spatial relationship between characters without fused vocabulary information and characters with fused vocabulary information, aiming to compensate for information loss and strengthen spatial connections. The comparison results with the baseline model show that the SISF method proposed in this paper improves the F1 metric by 0.72%, 3.12%, 1.07%, and 0.37% on the Resume, Weibo, Ontonotes 4.0, and MSRA datasets, respectively.

The proportion of juvenile wood affects the utilization of wood seriously, and the transition year of juvenile wood (JW) and mature wood (MW) plays a decisive role in the rotation and the modification of wood. To find out the demarcation of JW and MW, the tracheid length (TL) and microfibril angle (MFA) of early wood (EW) and late wood (LW) from four Chinese fir clones were measured by optical microscopy and X-ray diffraction. Then the data were analyzed by the k-means clustering method. The correlation and the differences among wood properties between JW and MW were compared. Results indicated that the LW showed better properties than that of EW, but the anatomical differences between EW and LW did not influence the demarcation of JW and MW. The cluster analysis of TL and MFA showed that the transition year was in the 16th year and the transition zone of EW and LW was different among clones. The MW has longer and wider tracheid, thicker cell walls, and smaller MFA. In terms of chemistry, MW had a higher content of holocellulose, α-cellulose, less content of extract, but no significant difference in lignin content compared with JW. The stabilization of chemical components was earlier than that of the anatomic properties. Correlation analysis showed that there were strong correlations between the chemical composition and anatomical characteristics in JW and MW. In general, compared with chemical components, anatomical indicators were more suitable for JW and MW demarcation. The differences and correlations between JW and MW properties provide a theoretical basis for wood rotation and planting.

In recent years, large-scale pretrained language models have become widely used in natural language processing tasks. On this basis, prompt learning has achieved excellent performance in specific few-shot classification scenarios. The core idea of prompt learning is to convert a downstream task into a masked language modelling task. However, different prompt templates can greatly affect the results, and finding an appropriate template is difficult and time-consuming. To this end, this study proposes a novel hybrid prompt approach, which combines discrete prompts and continuous prompts, to motivate the model to learn more semantic knowledge from a small number of training samples. By comparing the performance difference between discrete prompts and continuous prompts, we find that hybrid prompts achieve the best results, reaching a 73.82% F1 value in the test set. In addition, we analyze the effect of different virtual token lengths in continuous prompts and hybrid prompts in a few-shot cross-language topic classification scenario. The results demonstrate that there is a threshold for the length of virtual tokens, and too many virtual tokens decrease the performance of the model. It is better not to exceed the average length of the training set corpus. Finally, this paper designs a method based on vector similarity to explore the real meanings represented by virtual tokens. The experimental results show that the prompt automatically learnt from the virtual token has a certain correlation with the input text.