I am Hao Wu, currently a third-year master’s student in the Department of Computer Science at the University of Science and Technology of China (USTC). I am also a joint training student in the large model training group of the Machine Learning Platform Department at Tencent. My research interests are as follows:

• Spatio-temporal Prediction: With the guidance of Kun Wang and Yuxuan Liang, I explore foundational models for spatiotemporal data mining. Additionally, I am honored to collaborate with Xingjian Shi, whose ConvLSTM introduced me to this field.

• AI/ML for Science: Under the guidance of Xiao Luo, I study machine learning and fluid dynamics.

• Scaling Law: I am collaborating with my teammates to research the intersectional theory and applications of Scaling Law in large scientific computing models.

🔥 News

• 2025.0122:  🎉🎉 1 paper was accepted to ICLR2025 (Corresponding Author).
• 2024.1116:  🎉🎉 1 paper was accepted to KDD2025 ADS (First Author).
• 2024.0926:  🎉🎉 3 papers were accepted to NeurIPS2024 (First Author and Two co-author).
• 2024.0716:  🎉🎉 1 paper was accepted to ACM MM2024 (First Author).
• 2024.0517:  🎉🎉 1 papers was accepted to KDD2024 (First Author).
• 2024.0501:  🎉🎉 1 paper was accepted to ICML2024 (First Author).
• 2024.0116:  🎉🎉 1 paper was accepted to ICLR2024 (Spotlight) (Co-First Author).
• 2024.0221:  🎉🎉 1 paper was accepted to TKDE2024 (Co-First Author).
• 2023.1209:  🎉🎉 1 paper was accepted to AAAI2024 (First Author).
• 2023.0922:  🎉🎉 1 paper was accepted to NeurIPS2023 (Co-First Author).
• 2022.1009:  🎉🎉 National Scholarship, China, 2023 (top 0.1% nation-wide, From USTC).

📖 Research Experience

• Machine Learning Platform Department, Large model training group, Tencent
• 2023.08 - present, Research intern
• mentored by Jinbao Xue

• CityMind Lab, Hong Kong University of Science and Technology (Guangzhou)
• 2023.06 - 2024.06, Research intern
• Advisor Yuxuan Liang

📝 Publications

Spatio-temporal Prediction

• KDD2025 DynST: Dynamic Sparse Training for Resource-Constrained Spatio-Temporal Forecasting. Hao Wu, Haomin Wen, Guibin Zhang, Yutong Xia, Yuxuan Liang, Yu Zheng, Qingsong Wen, Kun Wang#. KDD, 2025.

• AAAI2024 Earthfarseer: versatile spatio-temporal dynamical systems modeling in one model. Hao Wu, Yuxuan Liang, Wei Xiong#, Zhengyang Zhou, Wei Huang, Shilong Wang, Kun Wang#. AAAI, 2024.

• ICLR2024 NuwaDynamics: Discovering and Updating in Causal Spatio-Temporal Modeling. Kun Wang^, Hao Wu^, Yifan Duan, Guibin Zhang, Kai Wang, Xiaojiang Peng, Yu Zheng, Yuxuan Liang#, Yang Wang#. ICLR, 2024.

• TKDE2024 Modeling spatio-temporal dynamical systems with neural discrete learning and levels-of-experts. Kun Wang^, Hao Wu^, Guibin Zhang, Junfeng Fang, Yuxuan Liang, Yuankai Wu, Roger Zimmermann, Yang Wang#. TKDE, 2024.

• TPAMI2024 Earthfarseer-V2: A Versatile All-in-One Model for Learning Complex Spatio-Temporal Dynamics. Hao Wu, Junfeng Fang, Yuxuan Liang, Guibin Zhang, Fan Xu, Wei Xiong, Qingsong Wen, Yu Zheng, Kun Wang#. TPAMI, 2024 (under review).

• ACM MM2024 PastNet: introducing physical inductive biases for spatio-temporal video prediction.Hao Wu, Wei Xiong, Fan Xu, Xiao Luo#, Chong Chen, Xian-Sheng Hua, Haixin Wang#. ACM MM, 2024.

• Arxiv BeamVQ: Aligning Space-Time Forecasting Model via Self-training on Physics-aware Metrics.Hao Wu, Xingjian Shi, Ziyue Huang, Penghao Zhao, Wei Xiong, Jinbao Xue, Yangyu Tao, Xiaomeng Huang, Weiyan Wang #. Arxiv.

Neural Operator Learning

• KDD2024 Neural Manifold Operators for Learning the Evolution of Physical Dynamics. Hao Wu, Kangyu Weng, Shuyi Zhou, Xiaomeng Huang#, Wei Xiong#. KDD, 2024.

• AI4TS(Oral) Neural Manifold Operator for Geophysical Fluid Dynamics Prediction. Wei Xiong, Kun Wang, Yuxuan Liang, Hao Wu#, Xiaomeng Huang#. AI for Time Series (AI4TS) Workshop @ AAAI, 2024.

Scientific Machine Learning

• Arxiv OneForecast: A Universal Framework for Global and Regional Weather Forecasting)). Yuan Gao, Hao Wu, Ruiqi Shu, huanshuo dong, Fan Xu, Rui Ray Chen, Yibo Yan, Qingsong Wen, Xuming Hu, Kun Wang, Jiahao Wu, Li Qing, Hui Xiong, Xiaomeng Huang#. Arxiv, 2025.

• ICLR2025 Open-CK: A Large Multi-Physics Fields Coupling benchmarks in Combustion Kinetics. Zaige Fei, Fan Xu, Junyuan Mao, Yuxuan Liang, Qingsong Wen, Kun Wang#, Hao Wu#, Yang Wang#. ICLR, 2024.

• NeurIPS2024 PURE: Prompt Evolution with Graph ODE for Out-of-distribution Fluid Dynamics Modeling. Hao Wu, Changhu Wang, Fan Xu, Jinbao Xue, Chong Chen, Xian-Sheng Hua, Xiao Luo#. NeurIPS, 2024.

• ICML2024 Prometheus: Out-of-distribution Fluid Dynamics Modeling with Disentangled Graph ODE. Hao Wu, Huiyuan Wang, Kun Wang, Weiyan Wang, ChanganYe, Yangyu Tao, Chong Chen, Xian-Sheng Hua, Xiao Luo#. ICML, 2024.

• Arxiv AI-GOMS: Large AI-Driven Global Ocean Modeling System. Wei Xiong, Yanfei Xiang, Hao Wu, Shuyi Zhou, Yuze Sun, Muyuan Ma, Xiaomeng Huang#. Arxiv, 2024.

• Arxiv Spatio-temporal fluid dynamics modeling via physical-awareness and parameter diffusion guidance. Hao Wu, Fan Xu, Yifan Duan, Ziwei Niu, Weiyan Wang, Gaofeng Lu, Kun Wang, Yuxuan Liang#, Yang Wang#. Arxiv, 2024.

Information Retrieval

• NeurIPS2023 IDEA: An Invariant Perspective for Efficient Domain Adaptive Image Retrieval. Haixin Wang^, Hao Wu^, Jinan Sun, Shikun Zhang, Chong Chen, Xian-Sheng Hua, Xiao Luo#. NeurIPS, 2023.

Selected Publications

OneForecast: A Universal Framework for Global and Regional Weather Forecasting

Yuan Gao, Hao Wu, Ruiqi Shu, huanshuo dong, Fan Xu, Rui Ray Chen, Yibo Yan, Qingsong Wen, Xuming Hu, Kun Wang, Jiahao Wu, Li Qing, Hui Xiong, Xiaomeng Huang#

Arxiv, 2025

(CCF None)

Open-CK: A Large Multi-Physics Fields Coupling benchmarks in Combustion Kinetics

Zaige Fei, Fan Xu, Junyuan Mao, Yuxuan Liang, Qingsong Wen, Kun Wang#, Hao Wu#, Yang Wang#

International Conference on Learning Representations (ICLR), 2025

(THU Rank A)

DynST: Dynamic Sparse Training for Resource-Constrained Spatio-Temporal Forecasting

Hao Wu, Haomin Wen, Guibin Zhang, Yutong Xia, Yuxuan Liang, Yu Zheng, Qingsong Wen, Kun Wang

Knowledge Discovery and Data Mining (KDD), 2025

(CCF Rank A)

Neural Manifold Operators for Learning the Evolution of Physical Dynamics

Hao Wu, Kangyu Weng, Shuyi Zhou, Xiaomeng Huang, Wei Xiong

Knowledge Discovery and Data Mining (KDD), 2024

(CCF Rank A)

Prometheus: Out-of-distribution Fluid Dynamics Modeling with Disentangled Graph ODE

Hao Wu, Huiyuan Wang, Kun Wang, Weiyan Wang, ChanganYe, Yangyu Tao, Chong Chen, Xian-Sheng Hua, Xiao Luo

The International Conference on Machine Learning (ICML), 2024

(CCF Rank A)

PastNet: Introducing Physical Inductive Biases for Spatio-temporal Video Prediction

Hao Wu, Fan Xu, Chong Chen, Xian-Sheng Hua, Xiao Luo, Haixin Wang

ACM Multimedia (MM), 2024

(CCF Rank A)

Earthfarsser: Versatile Spatio-Temporal Dynamical Systems Modeling in One Model

Hao Wu, Yuxuan Liang, Wei Xiong, Zhengyang Zhou, Wei Huang, Shilong Wang, Kun Wang

Association for the Advancement of Artificial Intelligence (AAAI), 2024

(CCF Rank A)

PURE: Prompt Evolution with Graph ODE for Out-of-distribution Fluid Dynamics Modeling

Hao Wu, Changhu Wang, Fan Xu, Jinbao Xue, Chong Chen, Xian-Sheng Hua, Xiao Luo

Conference on Neural Information Processing Systems (NeurIPS), 2024

(CCF Rank A)

NuwaDynamics: Discovering and Updating in Causal Spatio-Temporal Modeling

Kun Wang, Hao Wu, Yifan Duan, Guibin Zhang, Kai Wang, Xiaojiang Peng, Yu Zheng, Yuxuan Liang, Yang Wang

International Conference on Learning Representations (ICLR), 2024

(THU Rank A spotlight)

BeamVQ: Aligning Space-Time Forecasting Model via Self-training on Physics-aware Metrics

Hao Wu, Xingjian Shi, Ziyue Huang, Penghao Zhao, Wei Xiong, Jinbao Xue, Yangyu Tao, Xiaomeng Huang, Weiyan Wang

Arxiv

🎖 Honors and Awards

• 2024.09 First-class Academic Scholarship of the University of Science and Technology of China.
• 2022.10 National Scholarship, China (top 0.1% nation-wide).
• 2022.09 First-class Academic Scholarship of the University of Science and Technology of China.

💬 Invited Talks

• 2024.03, Application and Research of GNN in Meteorological Prediction. @ Sun Yat-sen University
• 2023.12, Earthfarseer: versatile spatio-temporal dynamical systems modeling in one model. @ AI TIME
• 2023.06, A Review of Spatio-Temporal Forecasting Models. @ Tsinghua University

💻 Academic service

• PC Member/Conference Reviewer: NeurIPS2023 Conference Reviewers, NeurIPS2024 Conference Reviewers, NeurIPS 2024 Datasets and Benchmarks Track Reviewers, ICLR 2024 Conference Reviewers, ICML 2024 Conference Reviewers, ACMMM 2024 Conference Reviewers, ICLR2025 Conference Reviewers, NeurIPS 2024 Datasets and Benchmarks Track Reviewers, AISTATS 2025 Conference Reviewers, AAAI 2025 Conference Program Committee

👨🏻 Miscellaneous

• 🏀 I am a big fan of basketball, i love Kobe Bryant and i like Fadeaway Shot. I also like Curry.
• 👑 I am very interested in history.