Probing the Pitfalls: Understanding SVD’s Shortcomings in Language Model Compression by Сергей Александрович Плетенев

“…Then, we applied low-rank factorization to its transformer layers using various singular value decomposition algorithms at different compression rates. After that, we used probing tasks to analyze the changes in the internal representations and linguistic knowledge of the compressed models. …”
Get full text

Defect Detection and 3D Reconstruction of Complex Urban Underground Pipeline Scenes for Sewer Robots by Ruihao Liu, Zhongxi Shao, Qiang Sun, Zhenzhong Yu

“…The Sewer-YOLO-Slim model achieves reductions of 60.2%, 60.0%, and 65.9% in model size, parameters, and floating-point operations (FLOPs), respectively, while improving the mean average precision (mAP) by 1.5%, reaching 93.5%. Notably, the pruned model is only 4.9 MB in size. Comprehensive comparisons and analyses are conducted with 12 mainstream detection algorithms to validate the superiority of the proposed model. …”
Get full text

Optimizing Beam Stiffness and Beam Modal Response with Variable Spacing and Extrusion (VaSE) by Patrick N. Murphy, Richard A. Vittum, Bashir Khoda

“…This paper presents a novel algorithm, Variable Spacing and Extrusion (VaSE), designed to optimize the infill pattern of material extrusion (ME) 3D-printed parts for specified mechanical performance while ensuring manufacturability. …”
Get full text

Boosting grapevine phenological stages prediction based on climatic data by pseudo-labeling approach by Mehdi Fasihi, Mirko Sodini, Alex Falcon, Francesco Degano, Paolo Sivilotti, Giuseppe Serra

“…To ensure the robustness of the proposed Pseudo-labelling strategy, we integrated it into eight machine-learning algorithms. We evaluated its performance across seven diverse datasets, each exhibiting varying percentages of missing values. …”
Get full text

Ensemble Transformer–Based Detection of Fake and AI–Generated News by Md. Ishraquzzaman, Mohammed Ashraful Islam Chowdhury, Shahreen Rahman, Riasat Khan

“…This work leverages advanced natural language processing, machine learning, and deep learning algorithms to effectively detect fake and AI–generated content. …”
Get full text

LDDFSF-YOLO11: A Lightweight Insulator Defect Detection Method Focusing on Small-Sized Features by Peng Shen, Keyu Mei, Huiqiong Cao, Yongxiang Zhao, Guoqing Zhang

“…And optimize the network structure through channel pruning technology to improve model lightweighting. …”
Get full text

A Bicriteria Approach Identifying Nondominated Portfolios by Javier Pereira, Broderick Crawford, Fernando Paredes, Ricardo Soto

“…Next, a sorting process is applied on the input data or information matrix, which is intended to prune nonfeasible solutions early in the constructive algorithm. …”
Get full text

Diversifying Multi-Head Attention in the Transformer Model by Nicholas Ampazis, Flora Sakketou

“…Recent studies have shown that, due to redundancy, some heads of the Transformer model can be pruned without diminishing the efficiency of the model. …”
Get full text

EPO Extension of Dispatching Rules to Minimize Effects of Time Uncertainty in Production Scheduling by Radosław Puka, Iwona Skalna, Jerzy Duda, Tomasz Derlecki

“…., it has no negative impact on their most important advantage over other production scheduling algorithms. The results obtained from the real data of a large international offset printing company with various machine and job heterogeneities confirm the effectiveness of the proposed EPO extension in producing schedules that are indeed robust to changes in production time.…”
Get full text

Local Mean Normalization of Microarray Element Signal Intensities across an Array Surface: Quality Control and Correction of Spatially Systematic Artifacts by Carlo Colantuoni, George Henry, Scott Zeger, Jonathan Pevsne

“…These include artifacts that can be introduced during the robotic printing, hybridization, washing, or imaging of microarrays. …”
Get full text

Advanced Methods for Identifying Counterfeit Currency: Using Deep Learning and Machine Learning by Nama'a Hamed, Fadwa Al Azzo

“…Using machine learning algorithms like Random Forest, Decision Tree Classifier, XGBoost, CatBoost, and Support Vector Machine (SVM) in addition to deep learning techniques like Convolutional Neural Networks (CNNs), VGG16, MobileNetV2, and InceptionV3, we examine the security characteristics of Iraqi dinar banknotes and build robust models. …”
Get full text

A Review of Generative Design Using Machine Learning for Additive Manufacturing by Parankush Koul

“…Generative design uses complex algorithms to automate the process of designing best-fit designs, mass customization, and customization to suit specific customer requirements with high efficiency and quality. …”
Get full text

Evaluation of Liver Fibrosis through Noninvasive Tests in Steatotic Liver Disease by Yuri Cho

Get full text

A study on plant root apex morphology as a model for soft robots moving in soil. by Anand Kumar Mishra, Francesca Tramacere, Roberto Guarino, Nicola Maria Pugno, Barbara Mazzolai

Get full text

TOPOLOGICAL IMAGE PROCESSING FOR COMPREHENSIVE DEFECT AND DEVIATION ANALYSIS USING ADAPTIVE BINARISATION by Igor Badanyuk, Igor Nevliudov, Dmytro Nikitin

“…The aim of the work is to improve the method of adaptive binarisation for images obtained by technical vision systems by developing an automatic algorithm for detecting the required value of the image binarisation window. …”
Get full text

Exploring the possibilities of MADDPG for UAV swarm control by simulating in Pac-Man environment by Artem Novikov, Sergiy Yakovlev, Ivan Gushchin

“…Future research should focus on comparing the effectiveness of MADDPG-trained models with multi-agent algorithms, such as Expectimax, Alpha-Beta Pruning, and Monte Carlo Tree Search (MCTS), to further understand the advantages and limitations of learning-based approaches compared with traditional decision-making methods in collaborative and adversarial UAV operations. …”
Get full text

An Effective One-shot Body Part Multi-View Reconstruction Device with Self-calibration Capabilities by M. Bonotto, M. Bonotto, D. Evangelista, M. Imperoli, A. Pretto

“…We employ a custom multi-view stereo reconstruction pipeline based on (Schönberger et al., 2016), which guarantees optimal results without the downsides of the supervised data-driven multi-view stereo algorithms, i.e. data collection and ground truth labeling. …”
Get full text

A Cross-Stage Focused Small Object Detection Network for Unmanned Aerial Vehicle Assisted Maritime Applications by Gege Ding, Jiayue Liu, Dongsheng Li, Xiaming Fu, Yucheng Zhou, Mingrui Zhang, Wantong Li, Yanjuan Wang, Chunxu Li, Xiongfei Geng

“…The CFSD-UAVNet model was evaluated on the publicly available SeaDronesSee maritime dataset and compared with other cutting-edge algorithms. The experimental results showed that the CFSD-UAVNet model achieved an mAP@50 of 80.1% with only 1.7 M parameters and a computational cost of 10.2 G, marking a 12.1% improvement over YOLOv8 and a 4.6% increase compared to DETR. …”
Get full text

Distributed variational sparse Bayesian compressed sensing based on factor graphs by Cui-tao ZHU, Fan YANG, Han-xin WANG, Zhong-jie LI

“…A distributed variational sparse Bayesian compressed spectrum sensing algorithm based on factor graph was proposed,which decomposed the global spectrum sensing problem into local problem based on factor and variation.Belief propagation was used for the statistical inference of the spectrum occupancy,to implement the “soft fusion”.The temporal and spatial correlation information providing two-dimensional redundancies was exchanged among cooperative cognitive users to improve the detection performance under low SNR.Meanwhile,the algorithm prunes the divergence of hyper-parameters and the corresponding basis functions for reducing the load of communication.The simulation results show that this method can effectively achieve performance of spectrum sensing under a low sampling rate and the low SNR.…”
Get full text

Recording dan Pencarian Silsilah Kambing Perah Menggunakan Algoritme Depth First Search by Rizki Wahyudi, Muhammad Iqbal, Lussy Dwi Oktaviana

“…To solve the problem, a Recording system is built for website-based livestock data logging and for genealogical search will use the Depth first search algorithm. In order for the research to be directed, this research uses waterfall method with stage requirement analysis, design, implementation, testing, and maintenance. …”
Get full text