Leyu Zheng¹, Mingming Xiao^1,*, Yi Ren², Ke Li¹, Chang Sun¹

CMC-Computers, Materials & Continua, Vol.86, No.3, 2026, DOI:10.32604/cmc.2025.072036 - 12 January 2026

Abstract In a wide range of engineering applications, complex constrained multi-objective optimization problems (CMOPs) present significant challenges, as the complexity of constraints often hampers algorithmic convergence and reduces population diversity. To address these challenges, we propose a novel algorithm named Constraint Intensity-Driven Evolutionary Multitasking (CIDEMT), which employs a two-stage, tri-task framework to dynamically integrates problem structure and knowledge transfer. In the first stage, three cooperative tasks are designed to explore the Constrained Pareto Front (CPF), the Unconstrained Pareto Front (UPF), and the ε-relaxed constraint boundary, respectively. A CPF-UPF relationship classifier is employed to construct a problem-type-aware… More >

Ahmet Ezgi¹, Aytuğ Onan^2,*

CMC-Computers, Materials & Continua, Vol.86, No.3, 2026, DOI:10.32604/cmc.2025.071301 - 12 January 2026

Abstract Knowledge distillation has become a standard technique for compressing large language models into efficient student models, but existing methods often struggle to balance prediction accuracy with explanation quality. Recent approaches such as Distilling Step-by-Step (DSbS) introduce explanation supervision, yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation. In this work, we propose a task-structured curriculum learning (TSCL) framework that structures training into three sequential phases: (i) prediction-only, to establish stable feature representations; (ii) joint prediction–explanation, to align task outputs with rationale generation; and (iii)… More >

Yuxuan Gu, Meng Wu^*, Qian Wang, Siguang Chen, Lijun Yang

CMC-Computers, Materials & Continua, Vol.75, No.1, pp. 493-512, 2023, DOI:10.32604/cmc.2023.035974 - 06 February 2023

Abstract In this paper, a deep learning-based method is proposed for crowd-counting problems. Specifically, by utilizing the convolution kernel density map, the ground truth is generated dynamically to enhance the feature-extracting ability of the generator model. Meanwhile, the “cross stage partial” module is integrated into congested scene recognition network (CSRNet) to obtain a lightweight network model. In addition, to compensate for the accuracy drop owing to the lightweight model, we take advantage of “structured knowledge transfer” to train the model in an end-to-end manner. It aims to accelerate the fitting speed and enhance the learning ability… More >

Chuanrong Wu^1,*, Haotian Cui¹, Zhi Lu², Xiaoming Yang³, Mark E. McMurtrey⁴

CMC-Computers, Materials & Continua, Vol.69, No.3, pp. 3275-3287, 2021, DOI:10.32604/cmc.2021.019969 - 24 August 2021

Abstract Big data knowledge, such as customer demands and consumer preferences, is among the crucial external knowledge that firms need for new product development in the big data environment. Prior research has focused on the profit of big data knowledge providers rather than the profit and pricing schemes of knowledge recipients. This research addresses this theoretical gap and uses theoretical and numerical analysis to compare the profitability of two pricing schemes commonly used by knowledge recipients: subscription pricing and pay-per-use pricing. We find that: (1) the subscription price of big data knowledge has no effect on… More >

Chuanrong Wu^1,*, Veronika Lee¹, Mark E. McMurtrey²

CMC-Computers, Materials & Continua, Vol.60, No.1, pp. 365-378, 2019, DOI:10.32604/cmc.2019.06949

Abstract Product innovation is regarded as a primary means for enterprises to maintain their competitive advantage. Knowledge transfer is a major way that enterprises access knowledge from the external environment for new product innovation. Knowledge transfer may face the risk of infringement of the intellectual property rights of other enterprises and the termination of licensing agreements by the knowledge source. Enterprises must develop independent innovation knowledge at the same time they profit from knowledge transfers. Therefore, new product development by an enterprise usually consists of three types of new knowledge: big data knowledge transferred from big… More >

Chuanrong Wu^{1, *}, Evgeniya Zapevalova¹, Yingwu Chen², Deming Zeng³, Francis Liu⁴

CMES-Computer Modeling in Engineering & Sciences, Vol.116, No.1, pp. 89-107, 2018, DOI:10.31614/cmes.2018.04041

Abstract With market competition becoming fiercer, enterprises must update their products by constantly assimilating new big data knowledge and private knowledge to maintain their market shares at different time points in the big data environment. Typically, there is mutual influence between each knowledge transfer if the time interval is not too long. It is necessary to study the problem of continuous knowledge transfer in the big data environment. Based on research on one-time knowledge transfer, a model of continuous knowledge transfer is presented, which can consider the interaction between knowledge transfer and determine the optimal knowledge More >

Chuanrong Wu^{1, *}, Evgeniya Zapevalova¹, Yingwu Chen², Feng Li³

CMC-Computers, Materials & Continua, Vol.54, No.3, pp. 269-285, 2018, DOI:10.3970/cmc.2018.054.269

Abstract In the big data environment, enterprises must constantly assimilate big data knowledge and private knowledge by multiple knowledge transfers to maintain their competitive advantage. The optimal time of knowledge transfer is one of the most important aspects to improve knowledge transfer efficiency. Based on the analysis of the complex characteristics of knowledge transfer in the big data environment, multiple knowledge transfers can be divided into two categories. One is the simultaneous transfer of various types of knowledge, and the other one is multiple knowledge transfers at different time points. Taking into consideration the influential factors, More >