KAFA-Merge: Model Merging Method With Layer-Based Bayes+Linear Search by Muzaffer Kaan Yuce, Mehmet Fatih Amasyali

“…In this paper, we consider the model merging process for large language models (LLMs) under a two-stage optimization framework. Traditional merging methods usually apply fixed blending rates to all layers, which ignores the different levels of contribution of layers to the output. …”
Get full text

Speed and Path Planning Optimization of Autonomous Vehicle to Minimize Lap Time in Racing Track by Young-Jin Roh, Ji-Ung Im, Jong-Hoon Won

“…This study proposes a trajectory optimization method for racing vehicles, aiming to maximize speed and path planning performance by estimating tire friction coefficients. …”
Get full text

Optimal scheduling method for multi-regional integrated energy system based on dynamic robust optimization algorithm and bi-level Stackelberg model by Bo Zhou, Erchao Li, Wenjing Liang

Get full text

Optimized methods for the targeted surveillance of extended-spectrum beta-lactamase-producing Escherichia coli in human stool by Sarah Gallichan, Sally Forrest, Esther Picton-Barlow, Claudia McKeown, Maria Moore, Eva Heinz, Nicholas A. Feasey, Joseph M. Lewis, Fabrice E. Graf

“…Here, we compared different steps in the microbiological workflow to determine the impact different pre-enrichment broths, pre-enrichment incubation times, selection in pre-enrichment, selective plating, and DNA extraction methods had on recovering ESBL-EC from human stool samples, with the aim to acquire high-quality DNA for sequencing and genomic epidemiology. …”
Get full text

Optimal DMD Koopman Data-Driven Control of a Worm Robot by Mehran Rahmani, Sangram Redkar

“…The dynamic mode decomposition (DMD) method is used to generate the Koopman operator. Finally, a linear quadratic regulator (LQR) control method is applied for the control of the worm robot. …”
Get full text

Illumination in Museums: Four-Primary White LEDs to Optimize the Protective Effect and Color Quality by Rui Dang, Nan Wang, Gang Liu, Ye Yuan, Jie Liu, Huijiao Tan

“…Our results can provide the theory and application basis to manufacture white LEDs that are suitable for illuminating op-TCPs; the method can be further used in preparing white LEDs for other cultural relics.…”
Get full text

Encrypted traffic classification method based on convolutional neural network by Rongna XIE, Zhuhong MA, Zongyu LI, Ye TIAN

“…Aiming at the problems of low accuracy, weak generality, and easy privacy violation of traditional encrypted network traffic classification methods, an encrypted traffic classification method based on convolutional neural network was proposed, which avoided relying on original traffic data and prevented overfitting of specific byte structure of the application.According to the data packet size and arrival time information of network traffic, a method to convert the original traffic into a two-dimensional picture was designed.Each cell in the histogram represented the number of packets with corresponding size that arrive at the corresponding time interval, avoiding reliance on packet payloads and privacy violations.The LeNet-5 convolutional neural network model was optimized to improve the classification accuracy.The inception module was embedded for multi-dimensional feature extraction and feature fusion.And the 1*1 convolution was used to control the feature dimension of the output.Besides, the average pooling layer and the convolutional layer were used to replace the fully connected layer to increase the calculation speed and avoid overfitting.The sliding window method was used in the object detection task, and each network unidirectional flow was divided into equal-sized blocks, ensuring that the blocks in the training set and the blocks in the test set in a single session do not overlap and expanding the dataset samples.The classification experiment results on the ISCX dataset show that for the application traffic classification task, the average accuracy rate reaches more than 95%.The comparative experimental results show that the traditional classification method has a significant decrease in accuracy or even fails when the types of training set and test set are different.However, the accuracy rate of the proposed method still reaches 89.2%, which proves that the method is universally suitable for encrypted traffic and non-encrypted traffic.All experiments are based on imbalanced datasets, and the experimental results may be further improved if balanced processing is performed.…”
Get full text

A, Evaluation methods for waterfront public spaces: insights from different spatial scales in chinese and international cities by Xiaowen Wu, Claudio Gambardella, Jiaqi Zhong

“…This research explores the evaluation methods for waterfront public spaces based on different project scales. …”
Get full text

Targets Spatial Coordinates Calculation in Angle-Difference Rangefinder Passive Location Complexes by Numerical Iterative Methods by A. A. Dmitrenko, S. Y. Sedyshev

Get full text

Evaluating the impact of sales methods on profitability in the Japanese software industry by Takaaki Ishikawa, Kiminori Gemba

Get full text

A new heuristic method of finding the initial basic feasible solution to solve the transportation problem by Bilqis Amaliah, Chastine Fatichah, Erma Suryani

“…The initial basic feasible solution (IBFS) is a significant step to achieve the minimal total cost (optimal solution) of the transportation problem. However, the existing methods of IBFS do not always provide a good feasible solution which can reduce the number of iterations to find the optimal solution. …”
Get full text

A Single-Field Finite Difference Time-Domain Method Verified Using a Novel Antenna Design with an Artificial Magnetic Conductor Enhanced Structure by Yongjun Qi, Weibo Liang, Yilan Hu, Liang Zhang, Cheng You, Yuxiang Zhang, Tianrun Yan, Hongxing Zheng

“…The Finite Difference Time-Domain (FDTD) method is a powerful tool for electromagnetic field analysis. …”
Get full text

Comparative effectiveness of different methods of using the analogue of gonadotropin-releasing hormone for ovulation stimulation in rabbit does by V. I. Koshevoy, I. O. Zhukova, S. V. Naumenko, O. A. Savichev

Get full text

New Findings on the Survival of <i>Durusdinium glynnii</i> Under Different Acclimation Methods to Low Salinities by Barbara de Cassia S. Brandão, Jéssika L. de Abreu, Deyvid Willame S. Oliveira, Clarissa Vilela F. da Silva Campos, Ingrid M. Thó de Aguiar, Pedro R. de Sena, Alfredo O. Gálvez, Carlos Yure B. Oliveira

Get full text

Two Modified QN-Algorithms for Solving Unconstrained Optimization Problems by Abbas Al-Bayati, Basim Hassan

Get full text

An artificial intelligence and machine learning-driven CFD simulation for optimizing thermal performance of blood-integrated ternary nano-fluid by Mohib Hussain, Du Lin, Hassan Waqas, Qasem M. Al-Mdallal

“…However, conventional methods for modelling and optimizing these frameworks frequently encounter challenges owing to their intricacy and the multitude of interconnected variables. …”
Get full text

Inertial-relaxed splitting for composite monotone inclusions by Oré, Ernesto, Mahey, Philippe, Ocaña, Eladio

Get full text

Computational machine learning estimation of digitoxin solubility in supercritical solvent at different temperatures utilizing ensemble methods by Hadil Faris Alotaibi, Waqed H. Hassan, Ahmed Kateb Jumaah Al-Nussairi, Narinderjit Singh Sawaran Singh, Ahmed Salah Al-Shati, M. M. Rekha, Subhashree Ray, Aashna Sinha, Gunjan Garg

Get full text

Different Methods for Using a Validated Comic to Promote the Safe Use of Fluoridated Toothpaste Among Children by Sérgio da Silva Lima, Luciane Zanin, Arlete Maria Gomes Oliveira, Flávia Martão Flório

Get full text

Different bee-flower survey methods impact conservation recommendations: comparing citizen science and academic surveys by Jaco J.T.C. Visser, Leon Marshall, Jiří Hadrava, Pieter Vanormelingen, Arthur Dielens, Jean-Marc Molenberg, Nicolas J. Vereecken

“…The bipartite networks differed substantially (compositional difference = 0.86), mainly due to rewiring of bee-flower interactions (0.69). …”
Get full text