Xuejin Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xuejin Chen has a total of 184 co-authors. Here are the top co-authors by number of shared publications:
- Zheng-Jun Zha
University of Science and Technology of China, Hefei, China
15 shared publications - Zhiwei Xiong
University of Science and Technology of China, Hefei, China
14 shared publications Feng Wu
Anhui Province Key Laboratory of Biomedical Imaging and Intelligent Processing, Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China
6 shared publications- Shaobo Min
Tencent Data Platform, Shenzhen, China
6 shared publications - Chaoqun Wang
School of Artificial Intelligence, South China Normal University, Guangzhou, China
5 shared publications - Jie Zhao
National Engineering Laboratory for Brain-inspired Intelligence Technology and Application, University of Science and Technology of China, Hefei, China
5 shared publications
Latest Publications
- Open Access
A medial entorhinal connectome reveals compartmentalized microcircuit architecture
The medial entorhinal cortex (MEC) transforms multimodal inputs into spatial representations, but the synaptic architecture underlying its computations remains unclear. Here we generated a synapse-resolution connectomic dataset spanning all six layers of mouse MEC and the adjacent parasubiculum with serial electron microscopy, providing a continuous volume for mapping cellular organization and connectivity across layers and regions. We found that clustered pyramidal neurons in MEC layer II extended oblique apical dendrites with a dorsolateral bias and long-range myelinated axons within the layer. Exhaustive reconstruction of one cluster revealed a central exclusion core (ExC) that spatially partitioned basal dendritic architecture and synaptic connectivity. A spatially constrained growth model reproduced the position-dependent dendritic patterns. Within the ExC, pyramidal neurons formed a compact recurrent excitatory network and converged onto shared inhibitory partners. Basket cells (BC) and recurrent Martinotti cells provided perisomatic and apical-dendritic feedback, respectively. Dense, compact reciprocal connectivity between L2P neurons and BCs within the ExC, together with BC axonal bands extending beyond the ExC, suggests a potential circuit mechanism for generating temporally structured inhibitory output from the pyramidal-cell cluster along directionally extended axons. Together, these findings link cellular-scale geometry and synaptic wiring to emergent circuit dynamics that extend from dense computation within individual clusters to region-wide modulation, providing a connectomic foundation for understanding the circuit basis of spatial navigation.
Open AccessA medial entorhinal connectome reveals compartmentalized microcircuit architecture
The medial entorhinal cortex (MEC) transforms multimodal inputs into spatial representations, but the synaptic architecture underlying its computations remains unclear. Here we generated a synapse-resolution connectomic dataset spanning all six layers of mouse MEC and the adjacent parasubiculum with serial electron microscopy, providing a continuous volume for mapping cellular organization and connectivity across layers and regions. We found that clustered pyramidal neurons in MEC layer II extended oblique apical dendrites with a dorsolateral bias and long-range myelinated axons within the layer. Exhaustive reconstruction of one cluster revealed a central exclusion core (ExC) that spatially partitioned basal dendritic architecture and synaptic connectivity. A spatially constrained growth model reproduced the position-dependent dendritic patterns. Within the ExC, pyramidal neurons formed a compact recurrent excitatory network and converged onto shared inhibitory partners. Basket cells (BC) and recurrent Martinotti cells provided perisomatic and apical-dendritic feedback, respectively. Dense, compact reciprocal connectivity between L2P neurons and BCs within the ExC, together with BC axonal bands extending beyond the ExC, suggests a potential circuit mechanism for generating temporally structured inhibitory output from the pyramidal-cell cluster along directionally extended axons. Together, these findings link cellular-scale geometry and synaptic wiring to emergent circuit dynamics that extend from dense computation within individual clusters to region-wide modulation, providing a connectomic foundation for understanding the circuit basis of spatial navigation.
- 1 October 2026
- Published by Springer Nature
- p. 636-653
MV-STRIDE: Enabling MLLMs to Master Multi-view Spatial Reasoning via Hierarchical Capability Modeling
- 1 October 2026
- Published by Springer Nature
- p. 636-653
MV-STRIDE: Enabling MLLMs to Master Multi-view Spatial Reasoning via Hierarchical Capability Modeling
- 2 August 2026
- Published by Springer Nature
- p. 380-392
MC-Gaussian: 3D Gaussian Splatting with Multi-camera Setup in Autonomous Driving
- 2 August 2026
- Published by Springer Nature
- p. 380-392
MC-Gaussian: 3D Gaussian Splatting with Multi-camera Setup in Autonomous Driving
- 2 August 2026
- Published by Springer Nature
- p. 393-404
Efficient Modeling of Long-Range Morphology for 3D Neuron Reconstruction from Electron Microscopy Images
- 2 August 2026
- Published by Springer Nature
- p. 393-404
Efficient Modeling of Long-Range Morphology for 3D Neuron Reconstruction from Electron Microscopy Images
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. PP (21682194) , 1-13
AnyMetric: Universal Metric Optimization for Medical Image Segmentation With Preference Learning
Medical image segmentation is crucial for accurate disease diagnosis and treatment. To assess segmentation performance, various evaluation metrics have been developed. Existing methods typically design surrogate loss functions tailored to specific metrics for model optimization. However, it is challenging and laborious to customize loss functions for all non-differentiable and complex evaluation metrics. In this paper, we propose Preference-Guided Segmentation Network (PGSNet), a novel framework that enables the unified optimization of arbitrary evaluation metrics, eliminating the need for handcrafted surrogate loss functions. To this end, PGSNet designs a preference-based optimization strategy, which optimizes the segmentation model by directly learning human-defined evaluation preferences rather than relying on loss functions. Specifically, PGSNet first generates diverse segmentation candidates by leveraging multi-scale representations that capture different levels of medical information. These candidates are then assessed and ranked using a reward function derived from target evaluation metrics, reflecting segmentation preference according to task metrics. Finally, segmentation preference optimization is developed to leverage these preference signals to guide model training, ensuring that the probability of generating superior segmentation results over that of inferior ones. Instead of designing specific surrogate loss functions, PGSNet offers a generalizable and flexible optimization framework capable of handling various metrics. Extensive experiments on several challenging public datasets demonstrate that PGSNet achieves superior segmentation results compared to existing methods.
- 10 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. PP (21682194) , 1-13
AnyMetric: Universal Metric Optimization for Medical Image Segmentation With Preference Learning
Medical image segmentation is crucial for accurate disease diagnosis and treatment. To assess segmentation performance, various evaluation metrics have been developed. Existing methods typically design surrogate loss functions tailored to specific metrics for model optimization. However, it is challenging and laborious to customize loss functions for all non-differentiable and complex evaluation metrics. In this paper, we propose Preference-Guided Segmentation Network (PGSNet), a novel framework that enables the unified optimization of arbitrary evaluation metrics, eliminating the need for handcrafted surrogate loss functions. To this end, PGSNet designs a preference-based optimization strategy, which optimizes the segmentation model by directly learning human-defined evaluation preferences rather than relying on loss functions. Specifically, PGSNet first generates diverse segmentation candidates by leveraging multi-scale representations that capture different levels of medical information. These candidates are then assessed and ranked using a reward function derived from target evaluation metrics, reflecting segmentation preference according to task metrics. Finally, segmentation preference optimization is developed to leverage these preference signals to guide model training, ensuring that the probability of generating superior segmentation results over that of inferior ones. Instead of designing specific surrogate loss functions, PGSNet offers a generalizable and flexible optimization framework capable of handling various metrics. Extensive experiments on several challenging public datasets demonstrate that PGSNet achieves superior segmentation results compared to existing methods.
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 5254-5261
Long Video Understanding With Learnable Retrieval in Video-Language Models
The remarkable natural language understanding, reasoning, and generation capabilities of large language models (LLMs) have made them attractive for application to video understanding, utilizing video tokens as contextual input. However, employing LLMs for long video understanding presents significant challenges. The extensive number of video tokens leads to considerable computational costs for LLMs while using aggregated tokens results in loss of vision details. Moreover, the presence of abundant question-irrelevant tokens introduces noise to the video reasoning process. To address these issues, we introduce a simple yet effective learnable retrieval-based video-language model (R-VLM) for efficient long video understanding. Specifically, given a question and a long video, our model identifies the most relevant $K$ video chunks and uses their associated visual tokens to serve as context for the LLM inference. This effectively reduces the number of video tokens, eliminates noise interference, and enhances system performance. We achieve this by incorporating a learnable lightweight MLP block to facilitate the efficient retrieval of question-relevant chunks, through the end-to-end training of our video-language model with a proposed soft matching loss. Experimental results on multiple zero-shot video question answering datasets validate the effectiveness of our framework for comprehending long videos.
- 16 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 5254-5261
Long Video Understanding With Learnable Retrieval in Video-Language Models
The remarkable natural language understanding, reasoning, and generation capabilities of large language models (LLMs) have made them attractive for application to video understanding, utilizing video tokens as contextual input. However, employing LLMs for long video understanding presents significant challenges. The extensive number of video tokens leads to considerable computational costs for LLMs while using aggregated tokens results in loss of vision details. Moreover, the presence of abundant question-irrelevant tokens introduces noise to the video reasoning process. To address these issues, we introduce a simple yet effective learnable retrieval-based video-language model (R-VLM) for efficient long video understanding. Specifically, given a question and a long video, our model identifies the most relevant $K$ video chunks and uses their associated visual tokens to serve as context for the LLM inference. This effectively reduces the number of video tokens, eliminates noise interference, and enhances system performance. We achieve this by incorporating a learnable lightweight MLP block to facilitate the efficient retrieval of question-relevant chunks, through the end-to-end training of our video-language model with a proposed soft matching loss. Experimental results on multiple zero-shot video question answering datasets validate the effectiveness of our framework for comprehending long videos.
- 26 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45 (5) , 2574-2587
Neuron Segment Connectivity Prediction With Multimodal Features for Connectomics
Reconstructing neurons from large electron microscopy (EM) datasets for connectomic analysis presents a significant challenge, particularly in segmenting neurons of complex morphologies. Previous deep learning-based neuron segmentation methods often rely on pixel-level image context and produce extensive oversegmented fragments. Detecting these split errors and merging the split neuron segments are non-trivial for various neurons in a large-scale EM data volume. In this work, we exploit multimodal features in the full workflow of automatic neuron proofreading. We propose a novel connection point detection network that utilizes both global 3D morphological features and high-resolution local image context to extract candidate segment pairs from massive adjacent segments. To effectively fuse the 3D morphological feature and the dense image features from very different scales, we design a proposal-based image feature sampling to improve the efficiency of multimodal cross-attentions. Integrating the connection point detection network with our connectivity prediction network which also utilizes multimodal features, we make a fully automatic neuron segment merging pipeline, closely imitating human proofreading. Comprehensive experimental results verify the effectiveness of the proposed modules and demonstrate the robustness of the entire pipeline in large-scale neuron reconstruction. The code and data are available at https://github.com/Levishery/ Neuron-Segment-Connection-Prediction.
- 26 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 45 (5) , 2574-2587
Neuron Segment Connectivity Prediction With Multimodal Features for Connectomics
Reconstructing neurons from large electron microscopy (EM) datasets for connectomic analysis presents a significant challenge, particularly in segmenting neurons of complex morphologies. Previous deep learning-based neuron segmentation methods often rely on pixel-level image context and produce extensive oversegmented fragments. Detecting these split errors and merging the split neuron segments are non-trivial for various neurons in a large-scale EM data volume. In this work, we exploit multimodal features in the full workflow of automatic neuron proofreading. We propose a novel connection point detection network that utilizes both global 3D morphological features and high-resolution local image context to extract candidate segment pairs from massive adjacent segments. To effectively fuse the 3D morphological feature and the dense image features from very different scales, we design a proposal-based image feature sampling to improve the efficiency of multimodal cross-attentions. Integrating the connection point detection network with our connectivity prediction network which also utilizes multimodal features, we make a fully automatic neuron segment merging pipeline, closely imitating human proofreading. Comprehensive experimental results verify the effectiveness of the proposed modules and demonstrate the robustness of the entire pipeline in large-scale neuron reconstruction. The code and data are available at https://github.com/Levishery/ Neuron-Segment-Connection-Prediction.
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. 32 (4) , 2898-2911
Hie4DGS: Hierarchical 4D Gaussian Splatting From Monocular Dynamic Video
Monocular dynamic video reconstruction is a typical ill-posed problem due to the limited observations and complex 3D motions. Despite the recent advances in dynamic 3D Gaussian splatting techniques, most of them still struggle with the monocular setting, since they heavily rely on geometric cues from multiple cameras or ignore the structural coherence among the optimized 3D Gaussains. To address this, we propose Hie4DGS, a novel hierarchical structure representation to model the complex dynamic motions from monocular dynamic videos. Specifically, we decompose the motions of a dynamic scene into groups of multiple structure granularities and progressively compose them to derive the motion of each 3D Gaussian. Building on this representation, we leverage hierarchical semantic segmentation to group Gaussians and initialize their motion using depth and tracking priors within each group. Additionally, we introduce a structure rendering loss that enforces consistency between the learned motion structure and semantic priors, further reducing motion ambiguity. Compared to the state-of-the-art dynamic Gaussian methods, we achieve significant improvement in rendering quality on monocular video datasets featuring complex real-world motions.
- 21 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. 32 (4) , 2898-2911
Hie4DGS: Hierarchical 4D Gaussian Splatting From Monocular Dynamic Video
Monocular dynamic video reconstruction is a typical ill-posed problem due to the limited observations and complex 3D motions. Despite the recent advances in dynamic 3D Gaussian splatting techniques, most of them still struggle with the monocular setting, since they heavily rely on geometric cues from multiple cameras or ignore the structural coherence among the optimized 3D Gaussains. To address this, we propose Hie4DGS, a novel hierarchical structure representation to model the complex dynamic motions from monocular dynamic videos. Specifically, we decompose the motions of a dynamic scene into groups of multiple structure granularities and progressively compose them to derive the motion of each 3D Gaussian. Building on this representation, we leverage hierarchical semantic segmentation to group Gaussians and initialize their motion using depth and tracking priors within each group. Additionally, we introduce a structure rendering loss that enforces consistency between the learned motion structure and semantic priors, further reducing motion ambiguity. Compared to the state-of-the-art dynamic Gaussian methods, we achieve significant improvement in rendering quality on monocular video datasets featuring complex real-world motions.
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9920-9929
CausalCtrl: Causality-Aware Control Framework for Text-Guided Visual Editing
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 9920-9929
Open AccessCausalCtrl: Causality-Aware Control Framework for Text-Guided Visual Editing