Tianyi Zhou
Visiting Assistant Professor of Natural Language Processing
Research Interests
Professor Zhou's teaching and research interests span machine learning, natural language processing, optimization, and multimodal generative AI. His recent works are dedicated to data selection/synthesis pipelines and curriculum for training AI models, mechanisms of memorization and generalization in generalist models, reasoning and alignment of large language models (LLMs) and multimodal models with humans, mixture-of-experts, and unified models (representation/understanding+generation, agent+world model, etc.). Email
Prior to joining MBZUAI, Professor Zhou was an assistant professor in the Department of Computer Science at the University of Maryland, College Park (UMD). He was also a research scientist at Google AI before joining UMD.
- Ph.D. in Computer Science, University of Washington, USA
- Best Student Paper Award at ICDM, 2013.
- Top 10% of high-scoring reviewers for NeurIPS, 2020.
A selection of Professor Zhou's research:
- Ziyue Li, Tianyi Zhou: "Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free." ICLR Oral (1.8%), 2025.
- Ming Li, Yanhong Li, Tianyi Zhou: "What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective." ACL Oral, 2025.
- Siyu Zhou, Tianyi Zhou, Yijun Yang, Guodong Long, Deheng Ye, Jing Jiang, and Chengqi Zhang: "WALL-E: World Alignment by NeuroSymbolic Learning improves World Model-based LLM Agents." NeurIPS, 2025.
- Zhongyang Li, Ziyue Li, Tianyi Zhou: "R2-T2: Re-Routing in Test-Time for Multimodal Mixture-of-Experts." ICML, 2025.
- Ziyue Li, Tian Li, Virginia Smith, Jeff Bilmes, Tianyi Zhou: "Many-Objective Multi-Solution Transport." ICLR, 2025.
- Yijun Liang*, Shweta Bhardwaj*, Tianyi Zhou: "Diffusion Curriculum: Synthetic-to-Real Generative Curriculum Learning via Image-Guided Diffusion." ICCV, 2025.
- Jiuhai Chen, Zhiyang Xu, Xichen Pan, Yushi Hu, Can Qin, Tom Goldstein, Lifu Huang, Tianyi Zhou, Saining Xie, Silvio Savarese, Le Xue, Caiming Xiong, Ran Xu: "BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Dataset." arXiv:2505.09568, 2025.
- Ziyue Li, Chenrui Fan, Tianyi Zhou: "Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test.", arXiv:2506.21551, 2025.
- Tianrui Guan*, Fuxiao Liu*, Xiyang Wu, Ruiqi Xian, Zongxia Li, Xiaoyu Liu, Xijun Wang, Lichang Chen, Furong Huang, Yaser Yacoob, Dinesh Manocha, and Tianyi Zhou: "HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models." CVPR, 2024.