Chun-Ming Xia
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Chun-Ming Xia has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Yu-Jie Xiong
School of Electronic and Electrical Engineering, Shanghai University of Engineering Science, Shanghai, China
4 shared publications Jia-Chen Zhang
School of Electronic and Electrical Engineering Shanghai University of Engineering Science 201620 Shanghai China
3 shared publicationsZheng Zhou
School of Electronic and Electrical Engineering Shanghai University of Engineering Science 201620 Shanghai China
3 shared publications- Xijiong Xie
School of Information Science and Engineering, Ningbo University, Ningbo, China
2 shared publications - Xiong-Ying Li
School of Economics, Guangdong University of Finance and Economics, Guangzhou, China
2 shared publications - Xue-Qi Lv
School of Materials Engineering, Shanghai University of Engineering Science, Shanghai 201620, China
2 shared publications
Latest Publications
- 5 January 2026
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 29 (1) , 22
CRA-U: lightweight U-Net with component ranking attention for skin lesion segmentation
- 5 January 2026
- Published by Springer Nature in Pattern Analysis and Applications
- Vol. 29 (1) , 22
CRA-U: lightweight U-Net with component ranking attention for skin lesion segmentation
- Open Access
Multi-Sample Anti-Aliasing and Constrained Optimization for 3D Gaussian Splatting
Recent advances in 3D Gaussian splatting have significantly improved real-time novel view synthesis, yet insufficient geometric constraints during scene optimization often result in blurred reconstructions of fine-grained details, particularly in regions with high-frequency textures and sharp discontinuities. To address this, we propose a comprehensive optimization framework integrating multisample anti-aliasing (MSAA) with dual geometric constraints. Our system computes pixel colors through adaptive blending of quadruple subsamples, effectively reducing aliasing artifacts in high-frequency components. The framework introduces two constraints: (a) an adaptive weighting strategy that prioritizes under-reconstructed regions through dynamic gradient analysis, and (b) gradient differential constraints enforcing geometric regularization at object boundaries. This targeted optimization enables the model to allocate computational resources preferentially to critical regions requiring refinement while maintaining global consistency. Extensive experimental evaluations across multiple benchmarks demonstrate that our method achieves state-of-the-art performance in detail preservation, particularly in preserving high-frequency textures and sharp discontinuities, while maintaining real-time rendering efficiency. Quantitative metrics and perceptual studies confirm statistically significant improvements over baseline approaches in both structural similarity (SSIM) and perceptual quality (LPIPS).
Open AccessMulti-Sample Anti-Aliasing and Constrained Optimization for 3D Gaussian Splatting
Recent advances in 3D Gaussian splatting have significantly improved real-time novel view synthesis, yet insufficient geometric constraints during scene optimization often result in blurred reconstructions of fine-grained details, particularly in regions with high-frequency textures and sharp discontinuities. To address this, we propose a comprehensive optimization framework integrating multisample anti-aliasing (MSAA) with dual geometric constraints. Our system computes pixel colors through adaptive blending of quadruple subsamples, effectively reducing aliasing artifacts in high-frequency components. The framework introduces two constraints: (a) an adaptive weighting strategy that prioritizes under-reconstructed regions through dynamic gradient analysis, and (b) gradient differential constraints enforcing geometric regularization at object boundaries. This targeted optimization enables the model to allocate computational resources preferentially to critical regions requiring refinement while maintaining global consistency. Extensive experimental evaluations across multiple benchmarks demonstrate that our method achieves state-of-the-art performance in detail preservation, particularly in preserving high-frequency textures and sharp discontinuities, while maintaining real-time rendering efficiency. Quantitative metrics and perceptual studies confirm statistically significant improvements over baseline approaches in both structural similarity (SSIM) and perceptual quality (LPIPS).
- Open Access
Wavelet Mixture of Experts for Time Series Forecasting
The field of time series forecasting is rapidly advancing, with recent large-scale Transformers and lightweight Multilayer Perceptron (MLP) models showing strong predictive performance. However, conventional Transformer models are often hindered by their large number of parameters and their limited ability to capture non-stationary features in data through smoothing. Similarly, MLP models struggle to manage multi-channel dependencies effectively. To address these limitations, we propose a novel, lightweight time series prediction model, WaveTS-B. This model combines wavelet transforms with MLP to capture both periodic and non-stationary characteristics of data in the wavelet domain. Building on this foundation, we propose a channel clustering strategy that incorporates a Mixture of Experts (MoE) framework, utilizing a gating mechanism and expert network to handle multi-channel dependencies efficiently. We propose WaveTS-M, an advanced model tailored for multi-channel time series prediction. Empirical evaluation across eight real-world time series datasets demonstrates that our WaveTS series models achieve state-of-the-art (SOTA) performance with significantly fewer parameters. Notably, WaveTS-M shows substantial improvements on multi-channel datasets, highlighting its effectiveness.
Open AccessWavelet Mixture of Experts for Time Series Forecasting
The field of time series forecasting is rapidly advancing, with recent large-scale Transformers and lightweight Multilayer Perceptron (MLP) models showing strong predictive performance. However, conventional Transformer models are often hindered by their large number of parameters and their limited ability to capture non-stationary features in data through smoothing. Similarly, MLP models struggle to manage multi-channel dependencies effectively. To address these limitations, we propose a novel, lightweight time series prediction model, WaveTS-B. This model combines wavelet transforms with MLP to capture both periodic and non-stationary characteristics of data in the wavelet domain. Building on this foundation, we propose a channel clustering strategy that incorporates a Mixture of Experts (MoE) framework, utilizing a gating mechanism and expert network to handle multi-channel dependencies efficiently. We propose WaveTS-M, an advanced model tailored for multi-channel time series prediction. Empirical evaluation across eight real-world time series datasets demonstrates that our WaveTS series models achieve state-of-the-art (SOTA) performance with significantly fewer parameters. Notably, WaveTS-M shows substantial improvements on multi-channel datasets, highlighting its effectiveness.
- Open Access
CausalDiffTab: Mixed-Type Causal-Aware Diffusion for Tabular Data Generation
Training data has been proven to be one of the most critical components in training generative AI. However, obtaining high-quality data remains challenging, with data privacy issues presenting a significant hurdle. To address the need for high-quality data. Synthesize data has emerged as a mainstream solution, demonstrating impressive performance in areas such as images, audio, and video. Generating mixed-type data, especially high-quality tabular data, still faces significant challenges. These primarily include its inherent heterogeneous data types, complex inter-variable relationships, and intricate column-wise distributions. In this paper, we introduce CausalDiffTab, a diffusion model-based generative model specifically designed to handle mixed tabular data containing both numerical and categorical features, while being more flexible in capturing complex interactions among variables. We further propose a hybrid adaptive causal regularization method based on the principle of Hierarchical Prior Fusion. This approach adaptively controls the weight of causal regularization, enhancing the model's performance without compromising its generative capabilities. Comprehensive experiments conducted on seven datasets demonstrate that CausalDiffTab outperforms baseline methods across all metrics. Our code is publicly available at: https://github.com/Godz-z/CausalDiffTab.
Open AccessCausalDiffTab: Mixed-Type Causal-Aware Diffusion for Tabular Data Generation
Training data has been proven to be one of the most critical components in training generative AI. However, obtaining high-quality data remains challenging, with data privacy issues presenting a significant hurdle. To address the need for high-quality data. Synthesize data has emerged as a mainstream solution, demonstrating impressive performance in areas such as images, audio, and video. Generating mixed-type data, especially high-quality tabular data, still faces significant challenges. These primarily include its inherent heterogeneous data types, complex inter-variable relationships, and intricate column-wise distributions. In this paper, we introduce CausalDiffTab, a diffusion model-based generative model specifically designed to handle mixed tabular data containing both numerical and categorical features, while being more flexible in capturing complex interactions among variables. We further propose a hybrid adaptive causal regularization method based on the principle of Hierarchical Prior Fusion. This approach adaptively controls the weight of causal regularization, enhancing the model's performance without compromising its generative capabilities. Comprehensive experiments conducted on seven datasets demonstrate that CausalDiffTab outperforms baseline methods across all metrics. Our code is publicly available at: https://github.com/Godz-z/CausalDiffTab.
- Open Access
Heterogeneous to Homogeneous Melting Transition Observed During a Single Process
The melting mechanism at medium heating rates is unclear, owing to the lack of accurate characterizations of structural changes in poly-directional melting conditions. Here, a particular multilayered nanostructure was selected to control the propagation of melting in a single direction. We predicted the heterogeneous to homogeneous melting (HeM to HoM) transition during a single melting process at medium heating rates of 10–400 K/ps by molecular dynamics (MD) simulations, without a change in heating rates. The information on structural changes for the HeM to HoM transition, including the loss of crystallinity and long- and short-range order, are clearly provided by both a single direction and the radial distribution functions. These results contribute to a more comprehensive understanding of the HeM to HoM transition induced by heating rates.
Open AccessHeterogeneous to Homogeneous Melting Transition Observed During a Single Process
The melting mechanism at medium heating rates is unclear, owing to the lack of accurate characterizations of structural changes in poly-directional melting conditions. Here, a particular multilayered nanostructure was selected to control the propagation of melting in a single direction. We predicted the heterogeneous to homogeneous melting (HeM to HoM) transition during a single melting process at medium heating rates of 10–400 K/ps by molecular dynamics (MD) simulations, without a change in heating rates. The information on structural changes for the HeM to HoM transition, including the loss of crystallinity and long- and short-range order, are clearly provided by both a single direction and the radial distribution functions. These results contribute to a more comprehensive understanding of the HeM to HoM transition induced by heating rates.
- Open Access
A Novel Tensile Fracture Location of Friction Plug Welding (FPW) Joints
The fracture position of a friction plug welding (FPW) joint is typically located at or near the thermo-mechanically affected zone (TMAZ). Here, we found that microcracks in all FPW specimens initiate at the deformed plug center (DPC) zone and then propagate through the plug center along 45° shear surfaces, because the lowest hardness occurs at the DPC zone rather than the TMAZ or other zones, and the DPC zone presents a tilt fiber-like microstructure. Such a tilt microstructure stimulates formations and deformations of microvoids and propagation of microcracks along 45° shear surfaces. The ultimate tensile strength (237.7 MPa) and yield strength (220.8 MPa) of the FPW joint reach 78.8% and 85.7% of the base metal, respectively. These results indicate that 6061-T6 aluminum alloy can be effectively joined by the FPW technique.
Open AccessA Novel Tensile Fracture Location of Friction Plug Welding (FPW) Joints
The fracture position of a friction plug welding (FPW) joint is typically located at or near the thermo-mechanically affected zone (TMAZ). Here, we found that microcracks in all FPW specimens initiate at the deformed plug center (DPC) zone and then propagate through the plug center along 45° shear surfaces, because the lowest hardness occurs at the DPC zone rather than the TMAZ or other zones, and the DPC zone presents a tilt fiber-like microstructure. Such a tilt microstructure stimulates formations and deformations of microvoids and propagation of microcracks along 45° shear surfaces. The ultimate tensile strength (237.7 MPa) and yield strength (220.8 MPa) of the FPW joint reach 78.8% and 85.7% of the base metal, respectively. These results indicate that 6061-T6 aluminum alloy can be effectively joined by the FPW technique.