publications

2026

1. 

   Vision to Geometry: 3D Spatial Memory for Sequential Embodied MLLM Reasoning and Exploration

   Zhongyi Cai#, Yi Du#, Chen Wang, and Yu Kong

   Under Review, 2026

   PDF

2025

1. 

   Open Set Face Forgery Detection via Dual-Level Evidence Collection

   Zhongyi Cai, Bryce Gernon, Wentao Bao, Yifan Li, Matthew Wright, and Yu Kong

   Under Review, 2025

   PDF Website

2. 

   IndustryEQA: Pushing the Frontiers of Embodied Question Answering in Industrial Scenarios

   Yifan Li#, Yuhang Chen#, Anh Dao#, lichi Li, Zhongyi Cai, Zhen Tan, Tianlong Chen, and Yu Kong

   Advances in Neural Information Processing Systems DB Track, 2025

   PDF Code

2024

1. 

   Understanding convergence and generalization in federated learning through feature learning theory

   Wei Huang, Ye Shi, Zhongyi Cai, and Taiji Suzuki

   In The Twelfth International Conference on Learning Representations, 2024

   Bib PDF

   @inproceedings{huang2023understanding,
     title = {Understanding convergence and generalization in federated learning through feature learning theory},
     author = {Huang, Wei and Shi, Ye and Cai, Zhongyi and Suzuki, Taiji},
     booktitle = {The Twelfth International Conference on Learning Representations},
     year = {2024},
   }

2023

1. 

   Fed-CO2: Cooperation of Online and Offline Models for Severe Data Heterogeneity in Federated Learning

   Zhongyi Cai, Ye Shi, Wei Huang, and Jingya Wang

   Advances in Neural Information Processing Systems, 2023

   Bib PDF Code Poster Slides

   @article{cai2024fed,
     title = {Fed-CO2: Cooperation of Online and Offline Models for Severe Data Heterogeneity in Federated Learning},
     author = {Cai, Zhongyi and Shi, Ye and Huang, Wei and Wang, Jingya},
     journal = {Advances in Neural Information Processing Systems},
     volume = {36},
     year = {2023},
   }

2. 

   Fedtp: Federated learning by transformer personalization

   Hongxia Li#, Zhongyi Cai#, Jingya Wang, Jiangnan Tang, Weiping Ding, Chin-Teng Lin, and Ye Shi

   IEEE transactions on neural networks and learning systems, 2023

   Bib PDF Code

   @article{li2023fedtp,
     title = {Fedtp: Federated learning by transformer personalization},
     author = {Li#, Hongxia and Cai#, Zhongyi and Wang, Jingya and Tang, Jiangnan and Ding, Weiping and Lin, Chin-Teng and Shi, Ye},
     journal = {IEEE transactions on neural networks and learning systems},
     year = {2023},
     publisher = {IEEE},
   }