An Wang
Email Google Scholar GitHub

An Wang 王安

Research Scientist · Tencent Hunyuan LLM Team, Beijing

I received my Ph.D. in Computer Science from Institute of Science Tokyo (formerly Tokyo Institute of Technology) in June 2025, advised by Prof. Naoaki Okazaki.


I am currently working at Tencent on the LLM Pretraining Team (Qingyun Project) of Hunyuan, as a core contributor to Hy3 (ranked among the top models by usage on OpenRouter since release), mainly responsible for model architecture design and base model production.


My research focuses on model architecture, training strategies, and scaling laws for large language models. I am also passionate about AI agents and automating complex workflows.

News
2026.04 Hy3-Preview released — Tencent Hunyuan's next-generation large language model.
2025.11 HMoE: Heterogeneous Mixture of Experts for Language Modeling received SAC Highlight Award at EMNLP 2025.
2025.11 Sparsifying Mamba accepted at EMNLP 2025 Findings.
2025.06 Received Ph.D. degree from Institute of Science Tokyo.
2025.05 Scaling Laws for Floating-Point Quantization Training accepted at ICML 2025.
2025 Excellence Award from the Association for Natural Language Processing (NLP 2025).
2024.05 Joined Tencent Hunyuan LLM Pretraining Team (Qingyun Project).
Selected Publications
EMNLP 2025
HMoE: Heterogeneous Mixture of Experts for Language Modeling
An Wang*, Xingwu Sun*, Ruobing Xie, Shuaipeng Li, Jiaqi Zhu, Zhen Yang, Pinxue Zhao, J.N. Han, Zhanhui Kang, Di Wang, Naoaki Okazaki, Cheng-zhong Xu
EMNLP 2025 Main ★ SAC Highlight Award
ICML 2025
Scaling Laws for Floating-Point Quantization Training
Xingwu Sun, Shuaipeng Li, Ruobing Xie, Weidong Han, Kan Wu, Zhen Yang, Yixing Li, An Wang, Shuai Li, Jinbao Xue, Yu Cheng, Yangyu Tao, Zhanhui Kang, Chengzhong Xu, Di Wang, Jie Jiang
ICML 2025
EMNLP 2025
Sparsifying Mamba
An Wang, Ruobing Xie, Shuaipeng Li, Xingwu Sun, Zhanhui Kang
EMNLP 2025 Findings
arXiv
Towards a Comprehensive Scaling Law of Mixture-of-Experts
Guoliang Zhao, Yuhan Fu, Shuaipeng Li, Xingwu Sun, Ruobing Xie, An Wang, Weidong Han, Zhen Yang, Weixuan Sun, Yudong Zhang, Cheng-zhong Xu, Di Wang, Jie Jiang
arXiv preprint, 2025
EACL 2023
DREEAM: Guiding Attention with Evidence for Improving Document-Level Relation Extraction
Youmi Ma, An Wang, Naoaki Okazaki
EACL 2023 · Cited by 161
arXiv
Self-Distillation for Multi-Token Prediction
Guoliang Zhao, Ruobing Xie, An Wang, Shuaipeng Li, Huaibing Xie, Xingwu Sun
arXiv preprint, 2026
arXiv
Lossless KV Cache Compression to 2%
Zhen Yang, J.N. Han, Kan Wu, Ruobing Xie, An Wang, Xingwu Sun, Zhanhui Kang
arXiv preprint, 2024
View all publications →
Education & Experience
Education
2022 – 25
Ph.D. Computer Science
Institute of Science Tokyo · Advised by Prof. Naoaki Okazaki
2020 – 22
M.Eng. Computer Science
Tokyo Institute of Technology · Advised by Prof. Haruo Yokota
2015 – 19
B.Eng. Mechatronic Engineering
Shanghai University
Experience
2024 –
Tencent
Hunyuan LLM Pretraining Team (Qingyun), Beijing
2023 – 24
NEC Laboratories Europe
Human-centric Team, Heidelberg, Germany
2019
Intsig
NLU Team, Shanghai
Honors & Awards
2025 SAC Highlight Award, EMNLP 2025
2025 Excellence Award, Association for Natural Language Processing
2023 Best Paper Award, 29th Annual Meeting of The Association for Natural Language Processing
2023 Spring Scholarship for Doctoral Students, Tokyo Institute of Technology
2022 Advanced Human Resource Development Fellowship, Tokyo Institute of Technology