Comprehensive Experiments Research Articles

Parents spontaneously scaffold the formation of conversational pacts with their children

AbstractAdults readily coordinate on temporary pacts about how to refer to things in conversation. Young children are also capable of forming pacts with peers given appropriate experimenter intervention. Here, we investigate whether parents may spontaneously provide a similar kind of scaffolding with U.S. children in a director–matcher task (N = 201, 49% female; ages 4, 6, 8). In Experiment 1, we show that parents initiate more clarification exchanges with younger children who, in turn, are more likely to adopt labels introduced by the parent. We then examine whether the benefit of such scaffolding acts primarily through childrens' difficulties with comprehension (Experiment 2) or production (Experiment 3). Our findings suggest that parents primarily scaffold pacts by easing children's production difficulties, modeling cooperative communication.

Cross-Granularity Infrared Image Segmentation Network for Nighttime Marine Observations

Infrared image segmentation in marine environments is crucial for enhancing nighttime observations and ensuring maritime safety. While recent advancements in deep learning have significantly improved segmentation accuracy, challenges remain due to nighttime marine scenes including low contrast and noise backgrounds. This paper introduces a cross-granularity infrared image segmentation network CGSegNet designed to address these challenges specifically for infrared images. The proposed method designs a hybrid feature framework with cross-granularity to enhance segmentation performance in complex water surface scenarios. To suppress feature semantic disparity against different feature granularity, we propose an adaptive multi-scale fusion module (AMF) that combines local granularity extraction with global context granularity. Additionally, incorporating a handcrafted histogram of oriented gradients (HOG) features, we designed a novel HOG feature fusion module to improve edge detection accuracy under low-contrast conditions. Comprehensive experiments conducted on the public infrared segmentation dataset demonstrate that our method outperforms state-of-the-art techniques, achieving superior segmentation results compared to professional infrared image segmentation methods. The results highlight the potential of our approach in facilitating accurate infrared image segmentation for nighttime marine observation, with implications for maritime safety and environmental monitoring.

Enhancing Text-to-Image Generation: Integrating CLIP and Diffusion Models for Improved Visual Accuracy and Semantic Consistency

Abstract. Text-to-Image (T2I) generation focuses on producing images that precisely match given textual descriptions by combining techniques from computer vision and natural language processing (NLP). Existing studies have shown an innovative approach to enhance T2I generation by integrating Contrastive Language-Image Pretraining (CLIP) embeddings with a Diffusion Model (DM). The method involves initially extracting rich and meaningful text embeddings using CLIP, which are then transformed into corresponding images. These images are progressively refined through an iterative denoising process enabled by diffusion models. Comprehensive experiments conducted on the MS-COCO dataset validate the proposed method, demonstrating significant improvements in image fidelity and the alignment between text and images. When compared to traditional models like Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), which often struggle with maintaining both visual quality and semantic accuracy, this hybrid model shows superior performance. Future research could explore optimizing hybrid models further and applying T2I technology to specialized fields, such as medical imaging and scientific visualization, expanding its potential use cases.

Semantic-Driven Crossmodal Fusion for Multimodal Sentiment Analysis

In multimodal sentiment analysis (MSA), the fusion strategies of multimodal features significantly influence the performance of MSA models. Previous works frequently face challenges in integrating heterogeneous data without fully leveraging the rich semantic content of text, resulting in poor information association. We propose an MSA model based on Text-Driven Crossmodal Fusion and Mutual Information Estimation, called TeD-MI, which comprises a Stacked Text-Driven Crossmodal Fusion (STDC) module, which efficiently fusions the three modalities driven by the text modality to optimize fusion feature representation and enhance semantic understanding. Furthermore, TeD-MI designed a mutual information estimation module to achieve the best balance between preserving task-related information and filtering out irrelevant noise information as much as possible. Comprehensive experiments conducted on the CMU-MOSI and CMU-MOSEI datasets demonstrate our proposed model achieves varying degrees of improvement across most evaluation metrics.

Dynamic graph structure and spatio-temporal representations in wind power forecasting

Wind Power Forecasting (WPF) has gained considerable focus as a crucial aspect of the successful integration and operation of wind power. However, due to the stochastic and unstable nature of wind, it poses a real challenge to effectively analyze the correlations among multiple time series data for accurate prediction. In our study, an end-to-end framework called Dynamic Graph structure and Spatio-Temporal representation learning (DSTG) framework is proposed to achieve stable power forecasting by constructing graph data to capture the critical features in the data. Specifically, a graph structure learning (GSL) module is introduced to dynamically construct task-related correlation matrices via backpropagation to mitigate the inherent inconsistency and randomness of wind power data. Additionally, a dual-scale temporal graph learning (DTG) module is further proposed to explore the implicit spatio-temporal features at a fine-grained level using different skip connections from the constructed graph data. Finally, comprehensive experiments are performed on the collected Xuji Group Wind Power (XGWP) dataset, and the results show that DSTG outperforms the state-of-the-art spatio-temporal methods by 10.12% on the average of root mean square error and mean absolute error, demonstrating the effectiveness of DSTG. In conclusion, our model provides a promising approach.

Data discretization impact on deep learning for missing value imputation of continuous data

In various fields of information examination, for example, AI, profoundly getting the hang of missing information is a typical issue. Missing qualities should be tended to since they can adversely affect the exactness and adequacy of prescient models. This research investigates how data discretization affects deep learning methods for filling the missing values in datasets with continuous features. They provide a unique method for imputing missing values using deep neural networks (DNNs) called extravagant expectation maximization-deep neural network (EEM-DNN). This approach discretizes continuous features into separate intervals initially. This is justified by treating the issue of missing value imputation as a classification work, with the missing values being considered a distinct class. A DNN, designed explicitly for imputation, is then trained using the discretized data. The expectation maximization concepts are incorporated into the network architecture, and as a result, the network iteratively improves its imputation predictions. They run comprehensive experiments on several datasets from different fields to gauge the efficacy of the suggested strategy. The effectiveness of EEM-DNN is compared to that of other imputation approaches, such as traditional imputation techniques and deep learning methods without data discretization. Our findings show that data discretization significantly enhances imputation accuracy. In terms of imputation accuracy and prediction performance on downstream tasks, the EEM-DNN method regularly performs better than alternative methods. It also examines if various discretization techniques affect the overall imputation process. They find that the trade-off between bias and variance in imputed data depends on the discretization method selected. This highlights the significance of choosing a suitable discretization approach depending on the unique properties of the dataset.

A NEW CONJUGATE GRADIENT METHOD BASED ON LOGISTIC MAPPING FOR UNCONSTRAINED OPTIMIZATION AND ITS APPLICATION IN REGRESSION ANALYSIS

The study tackles the critical need for efficient optimization techniques in unconstrained optimization problems, where conventional techniques often suffer from slow and inefficient convergence. There is still a need for algorithms that strike a balance between computational efficiency and robustness, despite advancements in gradient-based techniques. This work introduces a novel conjugate gradient algorithm based on the logistic mapping formula. As part of the methodology, descent conditions are established, and the suggested algorithm's global convergence properties are thoroughly examined. Comprehensive numerical experiments are used for empirical validation, and the new algorithm is compared to the Polak-Ribière-Polyak (PRP) algorithm. The suggested approach performs better than the PR algorithm, according to the results, and is more efficient since it needs fewer function evaluations and iterations to reach convergence. Furthermore, the usefulness of the suggested approach is demonstrated by its actual use in regression analysis, notably in the modelling of population estimates for the Kurdistan Region of Iraq. In contrast to conventional least squares techniques, the method maintains low relative error rates while producing accurate predictions. All things considered, this study presents the novel conjugate gradient algorithm as an effective tool for handling challenging optimisation problems in both theoretical and real-world contexts.

Dynamic-Max-Value ReLU Functions for Adversarially Robust Machine Learning Models

The proliferation of deep learning has transformed artificial intelligence, demonstrating prowess in domains such as image recognition, natural language processing, and robotics. Nonetheless, deep learning models are susceptible to adversarial examples, well-crafted inputs that can induce erroneous predictions, particularly in safety-critical contexts. Researchers actively pursue countermeasures such as adversarial training and robust optimization to fortify model resilience. This vulnerability is notably accentuated by the ubiquitous utilization of ReLU functions in deep learning models. A previous study proposed an innovative solution to mitigate this vulnerability, presenting a capped ReLU function tailored to bolster neural network robustness against adversarial examples. However, the approach had a scalability problem. To address this limitation, a series of comprehensive experiments are undertaken across diverse datasets, and we introduce the dynamic-max-value ReLU function to address the scalability problem.

Through a Controlled Quenching to Achieve a Good Combination of Mechanical Properties in Low‐Yield Ratio 900 MPa High‐Strength Low‐Alloy Steels

Two distinct heat treatments, that is, intercritical quenching combined with low‐temperature tempering (QT) to control the initial quenching temperature, and salt‐bath quenching combined with partitioning (Q&P) to regulate the final quenching temperature, were employed on high‐strength low‐alloy (HSLA) steels to achieve multiphase microstructures characterized by high strength, low yield ratio, and good impact toughness. Comprehensive experiments involving tension test, low‐temperature impact, microstructural observation, and in situ tension have been conducted to compare the microstructures and mechanical properties. It is found that both kinds of specimens can achieve a good match between high strength‐toughness and low yield ratio. The microstructures are composed of lath martensite with ferrite for QT specimen and tempered martensite with bainite for the Q&P specimen, a combination of “soft” and “hard” phases thus resulting in a yield ratio lower than 0.85 for 900 MPa low‐alloy steels. Notably, the Q&P specimen exhibited a markedly superior uniform elongation of 6.1% compared to the QT counterpart of 3.2%, a phenomenon attributed to the work hardening rate during deformation. A combination of strong and weak lath structures in tempered martensite and bainite can induce dislocation propagation and the carbides can act as obstacles to dislocation motion, jointly enhancing work hardening.

D‐Electrons of Platinum Alloy Steering CO Pathway for Low‐Charge Potential Li‐CO2 Batteries

AbstractAprotic Li‐CO2 batteries suffer from sluggish solid‐solid co‐oxidation kinetics of C and Li2CO3, requiring extremely high charging potentials and leading to serious side reactions and poor energy efficiency. Herein, we introduce a novel approach to address these challenges by modulating the reaction pathway with tailored Pt d‐electrons and develop an aprotic Li‐CO2 battery with CO and Li2CO3 as the main discharge products. Note that the gas‐solid co‐oxidation reaction between CO and Li2CO3 is both kinetically and thermodynamically more favorable. Consequently, the Li‐CO2 batteries with CoPt alloy‐supported on nitrogen‐doped carbon nanofiber (CoPt@NCNF) cathode exhibit a charging potential of 2.89 V at 50 μA cm−2, which is the lowest charging potential to date. Moreover, the CoPt@NCNF cathode also shows exceptional cycling stability (218 cycles at 50 μA cm−2) and high energy efficiency up to 74.6 %. Comprehensive experiments and theoretical calculations reveal that the lowered d‐band center of CoPt alloy effectively promotes CO desorption and inhibits further CO reduction to C. This work provides promising insights into developing efficient and CO‐selective Li‐CO2 batteries.

Asymmetric Large Kernel Distillation Network for efficient single image super-resolution

Recently, significant advancements have been made in the field of efficient single-image super-resolution, primarily driven by the innovative concept of information distillation. This method adeptly leverages multi-level features to facilitate high-resolution image reconstruction, allowing for enhanced detail and clarity. However, many existing approaches predominantly emphasize the enhancement of distilled features, often overlooking the critical aspect of improving the feature extraction capabilities of the distillation module itself. In this paper, we address this limitation by introducing an asymmetric large-kernel convolution design. By increasing the size of the convolution kernel, we expand the receptive field, which enables the model to more effectively capture long-range dependencies among image pixels. This enhancement significantly improves the model's perceptual ability, leading to more accurate reconstructions. To maintain a manageable level of model complexity, we adopt a lightweight architecture that employs asymmetric convolution techniques. Building on this foundation, we propose the Lightweight Asymmetric Large Kernel Distillation Network (ALKDNet). Comprehensive experiments conducted on five widely recognized benchmark datasets-Set5, Set14, BSD100, Urban100, and Manga109-indicate that ALKDNet not only preserves efficiency but also demonstrates performance enhancements relative to existing super-resolution methods. The average PSNR and SSIM values show improvements of 0.10 dB and 0.0013, respectively, thereby achieving state-of-the art performance.

A Cap-Optimized mRNA Encoding Multiepitope Antigen ESAT6 Induces Robust Cellular and Humoral Immune Responses Against Mycobacterium tuberculosis

Background/Objectives. Tuberculosis is a deadly bacterial disease and the second most common cause of death from monoinfectious diseases worldwide. Comprehensive measures taken by health authorities in various countries in recent decades have saved tens of millions of lives, but the number of new cases of this infection has been steadily increasing in the last few years and already exceeds 10 million new cases annually. The development of new vaccines against tuberculosis is a priority area in the prevention of new cases of the disease. mRNA vaccines have already shown high efficacy against COVID-19 and other viral infections and can currently be considered a promising field of antituberculosis vaccination. In our previous study, we assessed the immunogenicity and protective activity of several types of antituberculosis mRNA vaccines with different 5′ untranslated regions, but the efficacy of these vaccines was either comparable with or lower than that of BCG. Methods. Here, we conducted a comprehensive experiment to investigate the effects of cotranscriptional capping conditions and of cap structure on the magnitude of the mRNAs’ translation in HEK293T and DC2.4 cells. The most effective cap version was used to create an antituberculosis mRNA vaccine called mEpitope-ESAT6. Results and Conclusions. We compared immunogenicity and protective activity between mEpitope-ESAT6 and BCG and found that the vaccine with the new cap type is more immunogenic than BCG. Nonetheless, the increased immunogenicity did not enhance vaccine-induced protection. Thus, the incorporation of different cap analogs into mRNA allows to modulate the efficacy of mRNA vaccines.

Novel Approach in Vegetation Detection Using Multi-Scale Convolutional Neural Network

Vegetation segmentation plays a crucial role in accurately monitoring and analyzing vegetation cover, growth patterns, and changes over time, which in turn contributes to environmental studies, land management, and assessing the impact of climate change. This study explores the potential of a multi-scale convolutional neural network (MSCNN) design for object classification, specifically focusing on vegetation detection. The MSCNN is designed to integrate multi-scale feature extraction and attention mechanisms, enabling the model to capture both fine and coarse vegetation patterns effectively. Moreover, the MSCNN architecture integrates multiple convolutional layers with varying kernel sizes (3 × 3, 5 × 5, and 7 × 7), enabling the model to extract features at different scales, which is vital for identifying diverse vegetation patterns across various landscapes. Vegetation detection is demonstrated using three diverse datasets: the CamVid dataset, the FloodNet dataset, and the multispectral RIT-18 dataset. These datasets present a range of challenges, including variations in illumination, the presence of shadows, occlusion, scale differences, and cluttered backgrounds, which are common in real-world scenarios. The MSCNN architecture allows for the integration of information from multiple scales, facilitating the detection of diverse vegetation types under varying conditions. The performance of the proposed MSCNN method is rigorously evaluated and compared against state-of-the-art techniques in the field. Comprehensive experiments showcase the effectiveness of the approach, highlighting its robustness in accurately segmenting and classifying vegetation even in complex environments. The results indicate that the MSCNN design significantly outperforms traditional methods, achieving a remarkable global accuracy and boundary F1 score (BF score) of up to 98%. This superior performance underscores the MSCNN’s capability to enhance vegetation detection in imagery, making it a promising tool for applications in environmental monitoring and land use management.

LGSur‐Net: A Local Gaussian Surface Representation Network for Upsampling Highly Sparse Point Cloud

AbstractWe introduce LGSur‐Net, an end‐to‐end deep learning architecture, engineered for the upsampling of sparse point clouds. LGSur‐Net harnesses a trainable Gaussian local representation by positioning a series of Gaussian functions on an oriented plane, complemented by the optimization of individual covariance matrices. The integration of parametric factors allows for the encoding of the plane's rotational dynamics and Gaussian weightings into a linear transformation matrix. Then we extract the feature maps from the point cloud and its adjoining edges and learn the local Gaussian depictions to accurately model the shape's local geometry through an attention‐based network. The Gaussian representation's inherent high‐order continuity endows LGSur‐Net with the natural ability to predict surface normals and support upsampling to any specified resolution. Comprehensive experiments validate that LGSur‐Net efficiently learns from sparse data inputs, surpassing the performance of existing state‐of‐the‐art upsampling methods. Our code is publicly available at https://github.com/Rangiant5b72/LGSur-Net.

Robust estimation of the incubation period and the time of exposure using γ-divergence

Estimating the exposure time to single infectious pathogens and the associated incubation period, based on symptom onset data, is crucial for identifying infection sources and implementing public health interventions. However, data from rapid surveillance systems designed for early outbreak warning often come with outliers originated from individuals who were not directly exposed to the initial source of infection (i.e. tertiary and subsequent infection cases), making the estimation of exposure time challenging. To address this issue, this study uses a three-parameter lognormal distribution and proposes a new γ-divergence-based robust approach for estimating the parameter corresponding to exposure time with a tailored optimization procedure using the majorization-minimization algorithm, which ensures the monotonic decreasing property of the objective function. Comprehensive numerical experiments and real data analyses suggest that our method is superior to conventional methods in terms of bias, mean squared error, and coverage probability of 95% confidence intervals.

A hybrid approach using support vector machine rule-based system: detecting cyber threats in internet of things

While the proliferation of the Internet of Things (IoT) has revolutionized several industries, it has also created severe data security concerns. The security of these network devices and the dependability of IoT networks depend on efficient threat detection. Device heterogeneity, computing resource constraints, and the ever-changing nature of cyber threats are a few of the obstacles that make detecting cyber threats in IoT systems difficult. Complex threats often go undetected by conventional security measures, requiring more sophisticated, adaptive detection methods. Therefore, this study presents the Hybrid approach based on the Support Vector Machines Rule-Based Detection (HSVMR-D) method for an all-encompassing approach to identifying cyber threats to the IoT. The HSVMR-D employs SVM to categorize known and unknown threats using attributes acquired from IoT data. Identifying known attack signatures and patterns using rule-based approaches improves detection efficiency without retraining by adapting pre-trained models to new IoT contexts. Moreover, protecting vital infrastructure and sensitive data, HSVMR-D provides a thorough and adaptable solution to improve the security posture of IoT deployments. Comprehensive experiment analysis and simulation results compared to the baseline study have confirmed the efficiency of the proposed HSVMR-D. Furthermore, increased resilience to completely novel changing threats, fewer false positives, and improved accuracy in threat detection are all outcomes that show the proposed work outperforms others. The HSVMR-D approach is helpful where the primary objective is a secure environment in the Internet of Things (IoT) when resources are limited.

Transformer Based Ship Detector: An Improvement on Feature Map and Tiny Training Set

The exponential increment of commodity exchange has raised the need for maritime border security in recent years. One of the most critical tasks for naval border security is ship detection inside and outside the territorial sea. Conventionally, the task requires a substantial human workload. Fortunately, with the rapid growth of the digital camera and deep-learning technique, computer programs can handle object detection tasks well enough to replace human labor. Therefore, this paper studies how to apply recent state-of-the-art deep-learning networks to the ship detection task. We found that with a suitable number of object queries, the Deformable-DETR method will improve the performance compared to the state-of-the-art ship detector. Moreover, comprehensive experiments on different scale datasets prove that the technique can significantly improve the results when the training sample is limited. Last but not least, feature maps given by the method will focus well on key objects in the image.

Exploring the limits of hierarchical world models in reinforcement learning.

Hierarchical model-based reinforcement learning (HMBRL) aims to combine the sample efficiency of model-based reinforcement learning with the abstraction capability of hierarchical reinforcement learning. While HMBRL has great potential, the structural and conceptual complexities of current approaches make it challenging to extract general principles, hindering understanding and adaptation to new use cases, and thereby impeding the overall progress of the field. In this work we describe a novel HMBRL framework and evaluate it thoroughly. We construct hierarchical world models that simulate the environment at various levels of temporal abstraction. These models are used to train a stack of agents that communicate top-down by proposing goals to their subordinate agents. A significant focus of this study is the exploration of a static and environment agnostic temporal abstraction, which allows concurrent training of models and agents throughout the hierarchy. Unlike most goal-conditioned H(MB)RL approaches, it also leads to comparatively low dimensional abstract actions. Although our HMBRL approach did not outperform traditional methods in terms of final episode returns, it successfully facilitated decision-making across two levels of abstraction. A central challenge in enhancing our method's performance, as uncovered through comprehensive experimentation, is model exploitation on the abstract level of our world model stack. We provide an in depth examination of this issue, discussing its implications and suggesting directions for future research to overcome this challenge. By sharing these findings, we aim to contribute to the broader discourse on refining HMBRL methodologies.

UPGCN: User Perception-Guided Graph Convolutional Network for Multimodal Recommendation

To tackle the challenges of cold start and data sparsity in recommendation systems, an increasing number of researchers are integrating item features, resulting in the emergence of multimodal recommendation systems. Although graph convolutional network-based approaches have achieved significant success, they still face two limitations: (1) Users have different preferences for various types of features, but existing methods often treat these preferences equally or fail to specifically address this issue. (2) They do not effectively distinguish the similarity between different modality item features, overlook the unique characteristics of each type, and fail to fully exploit their complementarity. To solve these issues, we propose the user perception-guided graph convolutional network for multimodal recommendation (UPGCN). This model consists of two main parts: the user perception-guided representation enhancement module (UPEM) and the multimodal two-step enhanced fusion method, which are designed to capture user preferences for different modalities to enhance user representation. At the same time, by distinguishing the similarity between different modalities, the model filters out noise and fully leverages their complementarity to achieve more accurate item representations. We performed comprehensive experiments on the proposed model, and the results indicate that it outperforms other baseline models in recommendation performance, strongly demonstrating its effectiveness.

Synergistic integration of fragmented transportation networks: When do networks (not) synergize?

The transportation of information, goods, people, and other entities inevitably experiences fluctuations in demand and supply, which results in the underutilization of certain links within a transportation network while other links are fully utilized and unable to accommodate any additional demand. A promising strategy for leveraging these links is the collaborative integration of fragmented transportation networks to facilitate access between isolated nodes. We present a novel analytical framework for examining the conditions under which such network integration would enhance the overall transportation efficiency. We conducted comprehensive numerical experiments to investigate the influence of the network structure on the synergistic effects of network integration with a focus on the link survival ratio and dissimilarity between networks. The results showed that networks comprising nodes with relatively uniform importance benefit significantly from network integration whereas core–periphery networks are less likely to benefit. Based on our findings, we propose an indicator for predicting the potential for synergistic effects that is related to the growth rate of the giant component. This study contributes understanding of transportation network dynamics and offers practical insights for optimizing network integration strategies to leverage underutilized capacities and improve various types of transportation systems.