Jiaxing Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jiaxing Xu has a total of 70 co-authors. Here are the top co-authors by number of shared publications:
- Yiping Ke
College of Computing and Data Science, Nanyang Technological University, Singapore
14 shared publications - Qingtian Bian
College of Computing and Data Science, Nanyang Technological University, Singapore, Singapore
6 shared publications Mengcheng Lan
5 shared publicationsQingtian Bian
College of Computing and Data Science Nanyang Technological University
4 shared publications- Tieying Li
School of Computer Science and Engineering, Northeastern University, Shenyang, China
4 shared publications Wei Zhang
Cognitive Neuroimaging Centre and Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore
4 shared publications
Latest Publications
- 28 October 2026
- Published by Elsevier in Neural Networks
- Vol. 206
Directed homophily-aware graph neural network
- 28 October 2026
- Published by Elsevier in Neural Networks
- Vol. 206
Directed homophily-aware graph neural network
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (8) , 9486-9501
Text4Seg++: Advancing Image Segmentation via Generative Language Modeling
Multimodal Large Language Models (MLLMs) have shown exceptional capabilities in vision-language tasks. However, effectively integrating image segmentation into these models remains a significant challenge. In this work, we propose a novel text-as-mask paradigm that casts image segmentation as a text generation problem, eliminating the need for additional decoders and significantly simplifying the segmentation process. Our key innovation is semantic descriptors, a new textual representation of segmentation masks where each image patch is mapped to its corresponding text label. We first introduce image-wise semantic descriptors, a patch-aligned textual representation of segmentation masks that integrates naturally into the language modeling pipeline. To enhance efficiency, we introduce the Row-wise Run-Length Encoding (R-RLE), which compresses redundant text sequences, reducing the length of semantic descriptors by 74% and accelerating inference by $3\times$, without compromising performance. Building upon this, our initial framework Text4Seg achieves strong segmentation performance across a wide range of vision tasks. To further improve granularity and compactness, we propose box-wise semantic descriptors, which localizes regions of interest using bounding boxes and represents region masks via structured mask tokens called semantic bricks. This leads to our refined model, Text4Seg++, which formulates segmentation as a next-brick prediction task, combining precision, scalability, and generative efficiency. Comprehensive experiments on natural and remote sensing datasets show that Text4Seg++ consistently outperforms state-of-the-art models across diverse benchmarks without any task-specific fine-tuning, while remaining compatible with existing MLLM backbones. Our work highlights the effectiveness, scalability, and generalizability of text-driven image segmentation within the MLLM framework.
- 1 August 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 48 (8) , 9486-9501
Text4Seg++: Advancing Image Segmentation via Generative Language Modeling
Multimodal Large Language Models (MLLMs) have shown exceptional capabilities in vision-language tasks. However, effectively integrating image segmentation into these models remains a significant challenge. In this work, we propose a novel text-as-mask paradigm that casts image segmentation as a text generation problem, eliminating the need for additional decoders and significantly simplifying the segmentation process. Our key innovation is semantic descriptors, a new textual representation of segmentation masks where each image patch is mapped to its corresponding text label. We first introduce image-wise semantic descriptors, a patch-aligned textual representation of segmentation masks that integrates naturally into the language modeling pipeline. To enhance efficiency, we introduce the Row-wise Run-Length Encoding (R-RLE), which compresses redundant text sequences, reducing the length of semantic descriptors by 74% and accelerating inference by $3\times$, without compromising performance. Building upon this, our initial framework Text4Seg achieves strong segmentation performance across a wide range of vision tasks. To further improve granularity and compactness, we propose box-wise semantic descriptors, which localizes regions of interest using bounding boxes and represents region masks via structured mask tokens called semantic bricks. This leads to our refined model, Text4Seg++, which formulates segmentation as a next-brick prediction task, combining precision, scalability, and generative efficiency. Comprehensive experiments on natural and remote sensing datasets show that Text4Seg++ consistently outperforms state-of-the-art models across diverse benchmarks without any task-specific fine-tuning, while remaining compatible with existing MLLM backbones. Our work highlights the effectiveness, scalability, and generalizability of text-driven image segmentation within the MLLM framework.
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 356-365
Bridge Breaking for Graph Neural Networks
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 356-365
Open AccessBridge Breaking for Graph Neural Networks
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 6421-6430
Multi-Domain Enhancement via Residual Interwoven Transfer in Cross-Domain Sequential Recommendation
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 6421-6430
Open AccessMulti-Domain Enhancement via Residual Interwoven Transfer in Cross-Domain Sequential Recommendation
- 19 September 2025
- Published by Springer Nature
- p. 172-182
BrainPrompt: Multi-level Brain Prompt Enhancement for Neurological Condition Identification
- 19 September 2025
- Published by Springer Nature
- p. 172-182
BrainPrompt: Multi-level Brain Prompt Enhancement for Neurological Condition Identification
- 16 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (2) , 1568-1579
Multi-Atlas Brain Network Classification Through Consistency Distillation and Complementary Information Fusion
Brain network analysis plays a crucial role in identifying distinctive patterns associated with neurological disorders. Functional magnetic resonance imaging (fMRI) enables the construction of brain networks by analyzing correlations in blood-oxygen-level-dependent (BOLD) signals across different brain regions, known as regions of interest (ROIs). These networks are typically constructed using atlases that parcellate the brain based on various hypotheses of functional and anatomical divisions. However, there is no standard atlas for brain network classification, leading to limitations in detecting abnormalities in disorders. Recent methods leveraging multiple atlases fail to ensure consistency across atlases and lack effective ROI-level information exchange, limiting their efficacy. To address these challenges, we propose the Atlas-Integrated Distillation and Fusion network (AIDFusion), a novel framework designed to enhance brain network classification using fMRI data. AIDFusion introduces a disentangle Transformer to filter out inconsistent atlas-specific information and distill meaningful cross-atlas connections. Additionally, it enforces subject- and population-level consistency constraints to improve cross-atlas coherence. To further enhance feature integration, AIDFusion incorporates an inter-atlas message-passing mechanism that facilitates the fusion of complementary information across brain regions. We evaluate AIDFusion on four resting-state fMRI datasets encompassing different neurological disorders. Experimental results demonstrate its superior classification performance and computational efficiency compared to state-of-the-art methods. Furthermore, a case study highlights AIDFusion's ability to extract interpretable patterns that align with established neuroscience findings, reinforcing its potential as a robust tool for multi-atlas brain network analysis. The code is publicly available at https://github.com/AngusMonroe/AIDFusion.
- 16 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (2) , 1568-1579
Multi-Atlas Brain Network Classification Through Consistency Distillation and Complementary Information Fusion
Brain network analysis plays a crucial role in identifying distinctive patterns associated with neurological disorders. Functional magnetic resonance imaging (fMRI) enables the construction of brain networks by analyzing correlations in blood-oxygen-level-dependent (BOLD) signals across different brain regions, known as regions of interest (ROIs). These networks are typically constructed using atlases that parcellate the brain based on various hypotheses of functional and anatomical divisions. However, there is no standard atlas for brain network classification, leading to limitations in detecting abnormalities in disorders. Recent methods leveraging multiple atlases fail to ensure consistency across atlases and lack effective ROI-level information exchange, limiting their efficacy. To address these challenges, we propose the Atlas-Integrated Distillation and Fusion network (AIDFusion), a novel framework designed to enhance brain network classification using fMRI data. AIDFusion introduces a disentangle Transformer to filter out inconsistent atlas-specific information and distill meaningful cross-atlas connections. Additionally, it enforces subject- and population-level consistency constraints to improve cross-atlas coherence. To further enhance feature integration, AIDFusion incorporates an inter-atlas message-passing mechanism that facilitates the fusion of complementary information across brain regions. We evaluate AIDFusion on four resting-state fMRI datasets encompassing different neurological disorders. Experimental results demonstrate its superior classification performance and computational efficiency compared to state-of-the-art methods. Furthermore, a case study highlights AIDFusion's ability to extract interpretable patterns that align with established neuroscience findings, reinforcing its potential as a robust tool for multi-atlas brain network analysis. The code is publicly available at https://github.com/AngusMonroe/AIDFusion.
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1286-1295
Divergent Paths: Separating Homophilic and Heterophilic Learning for Enhanced Graph-level Representations
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 1286-1295
Open AccessDivergent Paths: Separating Homophilic and Heterophilic Learning for Enhanced Graph-level Representations
- Open Access
- 2 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Multi-View Graph Learning in Brain Network Analysis: A Survey
Exploring the complex structure of the human brain is essential for understanding its functions and diagnosing neurological disorders. With advancements in neuroimaging technologies, graph-based approaches have become a powerful tool for modeling the brain, where regions of interest (ROIs) are represented as nodes, and the functional or structural relationships between them are represented as edges. These graph representations provide valuable insights into brain connectivity and network organization, enabling more effective analyses of brain activity and structure. In recent years, multi-view graph learning has emerged as a promising framework for brain network analysis, as it facilitates the integration of multiple neuroimaging modalities or data views, offering a more comprehensive understanding of the brain. This approach is particularly valuable for capturing the intricate and multi-faceted nature of brain networks, which is critical for diagnosing complex neurological disorders. This survey reviews the application of multi-view graph learning in brain network analysis, highlighting the necessity and advantages of graph learning in this context. The main contributions of this survey are: (1) an overview of commonly used brain network datasets; (2) a systematic review of multi-view brain network analysis, focusing on three key paradigms: multi-modal, multi-parcellation, and multi-temporal; and (3) a discussion of open challenges, the role of graph learning in overcoming them, and potential future research directions.
- 2 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessMulti-View Graph Learning in Brain Network Analysis: A Survey
Exploring the complex structure of the human brain is essential for understanding its functions and diagnosing neurological disorders. With advancements in neuroimaging technologies, graph-based approaches have become a powerful tool for modeling the brain, where regions of interest (ROIs) are represented as nodes, and the functional or structural relationships between them are represented as edges. These graph representations provide valuable insights into brain connectivity and network organization, enabling more effective analyses of brain activity and structure. In recent years, multi-view graph learning has emerged as a promising framework for brain network analysis, as it facilitates the integration of multiple neuroimaging modalities or data views, offering a more comprehensive understanding of the brain. This approach is particularly valuable for capturing the intricate and multi-faceted nature of brain networks, which is critical for diagnosing complex neurological disorders. This survey reviews the application of multi-view graph learning in brain network analysis, highlighting the necessity and advantages of graph learning in this context. The main contributions of this survey are: (1) an overview of commonly used brain network datasets; (2) a systematic review of multi-view brain network analysis, focusing on three key paradigms: multi-modal, multi-parcellation, and multi-temporal; and (3) a discussion of open challenges, the role of graph learning in overcoming them, and potential future research directions.
- Open Access
$α$-GAN by Rényi Cross Entropy
This paper proposes $α$-GAN, a generative adversarial network using Rényi measures. The value function is formulated, by Rényi cross entropy, as an expected certainty measure incurred by the discriminator's soft decision as to where the sample is from, true population or the generator. The discriminator tries to maximize the Rényi certainty about sample source, while the generator wants to reduce it by injecting fake samples. This forms a min-max problem with the solution parameterized by the Rényi order $α$. This $α$-GAN reduces to vanilla GAN at $α= 1$, where the value function is exactly the binary cross entropy. The optimization of $α$-GAN is over probability (vector) space. It is shown that the gradient is exponentially enlarged when Rényi order is in the range $α\in (0,1)$. This makes convergence faster, which is verified by experimental results. A discussion shows that choosing $α\in (0,1)$ may be able to solve some common problems, e.g., vanishing gradient. A following observation reveals that this range has not been fully explored in the existing Rényi version GANs.Open Access$α$-GAN by Rényi Cross Entropy
This paper proposes $α$-GAN, a generative adversarial network using Rényi measures. The value function is formulated, by Rényi cross entropy, as an expected certainty measure incurred by the discriminator's soft decision as to where the sample is from, true population or the generator. The discriminator tries to maximize the Rényi certainty about sample source, while the generator wants to reduce it by injecting fake samples. This forms a min-max problem with the solution parameterized by the Rényi order $α$. This $α$-GAN reduces to vanilla GAN at $α= 1$, where the value function is exactly the binary cross entropy. The optimization of $α$-GAN is over probability (vector) space. It is shown that the gradient is exponentially enlarged when Rényi order is in the range $α\in (0,1)$. This makes convergence faster, which is verified by experimental results. A discussion shows that choosing $α\in (0,1)$ may be able to solve some common problems, e.g., vanishing gradient. A following observation reveals that this range has not been fully explored in the existing Rényi version GANs.