What's new

(QVGen) “QVGen: Pushing the limit of quantized video generative models,” preprint. [Paper]
(ICML 25) Three papers accepted by ICML 2025.
- “AdaWorld: Learning adaptable world models with latent actions” [Project Page]
- “HarmoniCa: Harmonizing training and inference for better feature cache in diffusion transformer acceleration” [Paper]
- “C2IQL: Constraint-conditioned implicit Q-learning for safe offline reinforcement learning”
(ICLR 25) Two papers accepted by ICLR 2025.
- “GI-GS: Global illumination decomposition on Gaussian splatting for inverse rendering” [Paper] [Project Page]
- “Exponential topology-enabled scalable communication in multi-agent reinforcement learning” [Paper]
(NAACL 25) “Graph neural network enhanced retrieval for question answering of large language models,” accepted by NAACL 2025. [Paper]
(AISTATS 25) “Reinforcement learning with intrinsically motivated feedback graph for lost-sales inventory control,” accepted by AISTATS 2025. [Paper]
(INFOCOM 25) “WRF-GS: Wireless radiation field reconstruction with 3D Gaussian splatting,” accepted by INFOCOM 2025. [Paper] [Code]
(AAAI 25) Two papers accepted by AAAI 2025.
- “CAMSIC: Content-aware masked image modeling transformer for stereo image compression” [Paper]
- “Learn how to query from unlabeled data streams in federated learning” [Paper]
(NeurIPS 24) Two papers accepted to NeurIPS 2024.
- “Vista: A generalizable driving world model with high fidelity and versatile controllability” [Paper] [Code] [Demo]
- “Kaleidoscope: Learnable masks for heterogeneous multi-agent reinforcement learning” [Paper] [Code]
(Nature Communications) Paper “Selective knowledge sharing for privacy-preserving federated distillation without a good teacher” was accepted by Nature Communications. [Paper]

Research Interests

At iComAI Lab, we pursue cutting-edge research at the intersection of communications technology and AI, with a particular focus on intelligent decision making and spatial perception. Our work bridges theoretical foundations with practical applications, addressing fundamental challenges in:

Generative AI, Foundation Models
Neural Data Representation and Compression
Reinforcement Learning
Safe and Efficient AI
O-RAN, Edge AI Systems

Selected Publications

GenAI, Foundation Models

S. Gao, S. Zhou, Y. Du, J. Zhang, and C. Gan, “AdaWorld: Learning adaptable world models with latent actions,” International Conference on Machine Learning (ICML), Vancouver, Canada, July 2025. (Acceptance Rate: 26.9%) [Project Page]
W. Sun, S. Chen, F. Liu, Z. Chen, Y. Duan, J. Zhang, and Y. Wang, “DimensionX: Create any 3D and 4D scenes from a single image with controllable video diffusion,” preprint. [Paper] [Code] [Project Page]
Z. Li, Q. Guo, J. Shao, L. Song, J. Bian, J. Zhang, and R. Wang, “Graph neural network enhanced retrieval for question answering of large language models,” The Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), Albuquerque, New Mexico, USA, April-May 2025. [Paper]
S. Gao, J. Yang, L. Chen, K. Chitta, Y. Qiu, A. Geiger, J. Zhang, and H. Li, “Vista: A generalizable driving world model with high fidelity and versatile controllability,” The Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, Dec. 2024. [Paper] [Code] [Demo] (Acceptance Rate: 25.8%)
J. Yang*, S. Gao*, Y. Qiu*, L. Chen, T. Li, B. Dai, K. Chitta, P. Wu, J. Zeng, J. Zhang, A. Geiger, Y. Qiao, and H. Li, “Generalized predictive model for autonomous driving,” IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, USA, Jun. 2024. (* equal contribution) (Acceptance Rate: 23.6%) (Highlight, Top 2.8%) [Paper]

Neural Data Representation and Compression

Z. Liu, Y. Hu, X. Zhang, J. Shao, Z. Lin, and J. Zhang, “Dynamics-aware Gaussian splatting streaming towards fast on-the-fly training for 4D reconstruction,” preprint. [Paper] [Project Page]
X. Zhang, Z. Liu, Y. Zhang, X. Ge, D. He, T. Xu, Y. Wang, Z. Lin, S. Yan, and J. Zhang, “MEGA: Memory-efficient 4D Gaussian splatting for dynamic scenes,” preprint. [Paper]
H. Chen, Z. Lin, and J. Zhang, “GI-GS: Global illumination decomposition on Gaussian splatting for inverse rendering,” International Conference on Learning Representations (ICLR), Singapore, April 2025. (Acceptance Rate: 32.08%) [Paper] [Project Page]
X. Zhang*, X. Ge*, T. Xu, D. He, Y. Wang, H. Qin, G. Lu, J. Geng, and J. Zhang, “GaussianImage: 1000 FPS image representation and compression by 2D Gaussian splatting,” European Conference on Computer Vision (ECCV), Milano, Italy, Sept.-Oct. 2024. (* equal contribution) (Acceptance Rate: 27.9%) [Paper] [Code]
X. Zhang, R. Yang, D. He, X. Ge, T. Xu, Y. Wang, H. Qin, and J. Zhang, “Boosting neural representations for videos with a conditional decoder,” IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, USA, Jun. 2024. (Acceptance Rate: 23.6%) (Highlight, Top 2.8%) [Paper] [Code]

Reinforcement Learning

Z. Liu, X. Li, and J. Zhang, “C2IQL: Constraint-conditioned implicit Q-learning for safe offline reinforcement learning,” International Conference on Machine Learning (ICML), Vancouver, Canada, July 2025. (Acceptance Rate: 26.9%)
Z. Liu, X. Li. S. Chen, G. Li, J. Jiang, and J. Zhang, “Reinforcement learning with intrinsically motivated feedback graph for lost-sales inventory control,” International Conference on Artificial Intelligence and Statistics (AISTATS), Mai Khao, Thailand, May 2025. (Acceptance Rate: 31.3%) [Paper]
X. Li, X. Wang, C. Bai, and J. Zhang, “Exponential topology-enabled scalable communication in multi-agent reinforcement learning,” International Conference on Learning Representations (ICLR), Singapore, April 2025. (Acceptance Rate: 32.08%) [Paper]
X. Li, L. Pan, and J. Zhang, “Kaleidoscope: Learnable masks for heterogeneous multi-agent reinforcement learning,” The Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, Dec. 2024. (Acceptance Rate: 25.8%) [Paper] [Code]
X. Li, Z. Liu, S. Chen, and J. Zhang, “Individual contributions as intrinsic exploration scaffolds for multi-agent reinforcement learning,” International Conference on Machine Learning (ICML), Vienna, Austria, July 2024. (Acceptance Rate: 27.5%) [Paper] [Code] [Video]

Efficient and Safe AI

Y. Huang, R. Gong, J. Liu, Y. Ding, C. Lv, H. Qin, and J. Zhang, “QVGen: Pushing the limit of quantized video generative models,” preprint. [Paper]
Y. Huang, Z. Wang, R. Gong, J. Liu, X. Zhang, J. Guo, X. Liu, and J. Zhang, “HarmoniCa: Harmonizing training and inference for better feature cache in diffusion transformer acceleration,” International Conference on Machine Learning (ICML), Vancouver, Canada, July 2025. (Acceptance Rate: 26.9%) [Paper]
Y. Sun, X. Li, T. Lin, and J. Zhang, “Learn how to query from unlabeled data streams in federated learning,” Pro. AAAI Conference on Artificial Intelligence, Philadelphia, Pennsylvania, USA, Feb.-Mar. 2025. (Acceptance Rate: 23.4%) [Paper]
J. Shao, F. Wu, and J. Zhang, “Selective knowledge sharing for privacy-preserving federated distillation without a good teacher,” Nature Communications, vol. 15, no. 349, Jan 2024. [Paper]
B. Li*, Y. Shen*, J. Yang, Y. Wang, J. Ren, T. Che, J. Zhang, and Z. Liu, “Sparse Mixture-of-Experts are Domain Generalizable Learners,” International Conference on Learning Representations (ICLR), Kigali, Rwanda, May 2023. (Oral presentation) [Paper] [GitHub]
J. Shao, Y. Sun, S. Li, and J. Zhang, “DReS-FL: Dropout-resilient secure federated learning for non-IID clients via secret data sharing,” NeurIPS 2022. [Paper] (Acceptance Rate: 25.6%)

O-RAN, Edge AI

B. Liu, J. Tong, and J. Zhang, “LLM-Slice: Dedicated wireless network slicing for large language models,” The 22nd ACM Conference on Embedded Networked Sensor Systems - Posters and Demos, Hangzhou, China, Nov. 2024. [Paper]
Y. Mao, X. Yu, K. Huang, A.-Y. Zhang, and J. Zhang, “Green edge AI: A contemporary survey,” Proc. IEEE, vol. 112, no. 7, pp. 880-911, July 2024. [Paper]
J. Shao, X. Zhang, and J. Zhang, “Task-oriented communication for edge video analytics,” IEEE Trans. Wireless Commun., vol. 23, no. 5, pp. 4141-4154, May 2024. [Paper] [GitHub]
J. Shao, Y. Mao, and J. Zhang, “Task-oriented communication for multidevice cooperative edge inference,” IEEE Trans. Wireless Commun., vol. 11, no. 1, pp. 73-87, Jan. 2023. [Paper]
J. Shao, Y. Mao, and J. Zhang, “Learning task-oriented communication for edge inference: An information bottleneck approach,” IEEE J. Select. Areas Commun, vol. 40, no. 1, pp. 197-211, Jan. 2022. [Paper] [GitHub]

AI for Communications

W. Yu, H. He, S. Song, J. Zhang, L. Dai, L. Zheng, and K. B. Letaief, “AI and deep learning for THz ultra-massive MIMO: From model-driven approaches to foundation models,” preprint. [Paper]
C. Wen, J. Tong, Z. Lin, and J. Zhang, “WRF-GS: Wireless radiation field reconstruction with 3D Gaussian splatting,” Proc. IEEE INFOCOM, London, United Kingdom, May 2025. (Acceptance Rate: 18.7%) [Paper] [Code]
J. Tong, J. Shao, Q. Wu, W. Guo, Z. Li, Z. Lin, and J. Zhang, “WirelessAgent: Large language model agents for intelligent wireless networks,” submitted. [Paper]
J. Shao, J. Tong, Q. Wu, W. Guo, Z. Li, Z. Lin, and Jun Zhang, “WirelessLLM: Empowering large language models towards wireless intelligence,” Journal of Communications and Information Networks, vol. 9, no. 2, pp. 99-112, June 2024. [Paper]
W. Yu, Y. Shen, H. He, X. Yu, S.H. Song, J. Zhang, and K. B. Letaief, “An adaptive and robust deep learning framework for THz ultra-massive MIMO channel estimation,” IEEE J. Sel. Topics Signal Process., vol. 17, no. 4, pp. 761-776, July 2023. [Paper] [GitHub]
Y. Shen, Y. Shi, J. Zhang, and K. B. Letaief, “Graph neural networks for scalable radio resource management: architecture design and theoretical analysis,” IEEE J. Select. Areas Commun, vol. 39, no. 1, pp. 101–115, Jan. 2021. [Paper]

Jun Zhang

What's new

Research Interests

Selected Publications