Expert Merging in Sparse Mixture of Experts with Nash Bargaining
Published in arXiv preprint, 2025
NAMEx uses Nash bargaining and complex momentum to merge experts more fairly and efficiently, outperforming prior methods across tasks.
Publications
Journal Articles
Image-based machine learning quantitative evaluation of bead-cell binding interaction
Published in Sensors and Actuators A: Physical, 2025
Proposing a machine learning-based platform for analyzing bead-cell interactions with A549 lung cancer cells.
Development of an Autonomous Component Testing System with Reliability Improvement Using Computer Vision and Machine Learning
Published in Transactions on Computer and Information Technology (ECTI-CIT), 2024
This paper proposes SIFT-SVM as an effective method for automated camera-module inspection, demonstrating superior performance and practical viability over existing computer vision approaches in real manufacturing conditions.
Conference Papers
Magnetic Bead Conjugated Lung Tumor Cell Binding Efficiency Assessment Based on Deep-Learning Approach
Published in 2023 1st International Conference on Health Science and Technology (ICHST), 2023
This paper introduces an integrated deep learning approach that automates and accurately quantifies magnetic bead–cell binding, offering a reliable signal readout for CTC detection systems.