Publications

2026

1. EviGDA

   

   Graph Domain Adaptation Does Not End with Representation Learning

   Yongxue Xu^*, Ziqian Liu^*, Enze Zhang, and Maolin Wang^†

   Preprint

   An evidence-augmented graph domain adaptation framework that combines independently trained graph-aware and feature-only experts with entropy-aware alignment and post-training probability fusion.

   Bib

   @article{xu2026evigda,
     title={Graph Domain Adaptation Does Not End with Representation Learning},
     author={Xu, Yongxue and Liu, Ziqian and Zhang, Enze and Wang, Maolin},
     journal={Preprint},
     note={Equal contribution: Yongxue Xu and Ziqian Liu. Corresponding author: Maolin Wang},
     year={2026}
   }

2. POSReasoner

   

   Learning to Reason with Persistent Object States for Video Instance Segmentation

   Yongxue Xu^*, Boxue Yang^*, Ziqian Liu, Shaoqiu Zhang, and Linfeng Zhang^†

   Preprint

   A plug-and-play framework that reasons over persistent object states to preserve identities through occlusion, reappearance, and interactions between similar instances in long-term video segmentation.

   Bib

   @article{xu2026posreasoner,
     title={Learning to Reason with Persistent Object States for Video Instance Segmentation},
     author={Xu, Yongxue and Yang, Boxue and Liu, Ziqian and Zhang, Shaoqiu and Zhang, Linfeng},
     journal={Preprint},
     note={Equal contribution: Yongxue Xu and Boxue Yang. Corresponding author: Linfeng Zhang},
     year={2026}
   }

3. TNSE

   

   Solving the Robust Influence Maximization Problem in Competitive Multilayer Networks via a Diffusion-Aware Role-Guided Evolutionary Approach

   Yongxue Xu, Ziqian Liu, Yongqing Huang, and Shuai Wang^†

   IEEE Transactions on Network Science and Engineering

   Under review

   A diffusion-aware role-guided evolutionary framework for robust competitive influence maximization on heterogeneous multilayer networks.

   Bib

   @article{xu2026influence,
     title={Solving the Robust Influence Maximization Problem in Competitive Multilayer Networks via a Diffusion-Aware Role-Guided Evolutionary Approach},
     author={Xu, Yongxue and Liu, Ziqian and Huang, Yongqing and Wang, Shuai},
     journal={Under review at IEEE Transactions on Network Science and Engineering},
     note={Corresponding author: Shuai Wang},
     year={2026}
   }

4. NeurIPS

   

   LMM-Track4D: Eliciting 4D Dynamic Reasoning in LMMs via Trajectory-Grounded Dialogue

   Chaoyue Li^*, Yongxue Xu^*, Jie Feng^*, and Jiayu Ding^†

   arXiv preprint

   Submitted to NeurIPS 2026

   A trajectory-grounded dialogue benchmark and LMM framework for 4D dynamic reasoning, combining RTGE, a streaming TRK state token, and an OSK-RA decoder for stable 3D trajectory estimation under occlusion and viewpoint variation.

   arXiv Bib Code

   @article{li2026lmmtrack4d,
     title={LMM-Track4D: Eliciting 4D Dynamic Reasoning in LMMs via Trajectory-Grounded Dialogue},
     author={Li, Chaoyue and Xu, Yongxue and Feng, Jie and Ding, Jiayu},
     journal={arXiv preprint arXiv:2605.19390},
     note={Equal contribution: Chaoyue Li, Yongxue Xu, and Jie Feng. Corresponding author: Jiayu Ding},
     year={2026}
   }

2025

1. GBCESC

   

   Identifying Critical Nodes with Deep Learning and Reinforcement Learning: A Case Study on Urban Road Networks

   Zhonghan Liu, Wenshuo Liu, Yongxue Xu, Ziqian Liu, Yichen Liu, and Shuai Wang^†

   GBCESC 2025

   Best Paper Award

   A benchmark of deep learning and reinforcement learning methods for critical node identification on real urban road networks.

   Bib

   @inproceedings{liu2025criticalnodes,
     title={Identifying Critical Nodes with Deep Learning and Reinforcement Learning: A Case Study on Urban Road Networks},
     author={Liu, Zhonghan and Liu, Wenshuo and Xu, Yongxue and Liu, Ziqian and Liu, Yichen and Wang, Shuai},
     booktitle={GBCESC},
     note={Best Paper Award. Corresponding author: Shuai Wang},
     year={2025}
   }