Shuo Wang is an Assistant Researcher / Assistant Professor at the Institute for AI Industry Research (AIR), Tsinghua University. Shuo Wang’s research interests include natural language processing, AI for healthcare, and AI for Science. He has published more than 40 papers at leading international conferences, including NeurIPS, ICLR, and ACL, and has received over 3,000 citations on Google Scholar. His doctoral research was recognized by the Doctoral Dissertation Incentive Program of the Chinese Information Processing Society of China. His recent work focuses on computational efficiency bottlenecks in large language models for long-horizon tasks. He has proposed several innovative solutions: Delta-CoMe, a mixed-precision quantization algorithm that enables efficient loading of up to 50 models with 7 billion parameters each on a single GPU; LLM×MapReduce, a divide-and-conquer framework for long-context processing that enables a 4-billion-parameter model to achieve performance comparable to contemporary flagship closed-source models on long-context tasks; and A\*-Thought, a chain-of-thought compression algorithm that uses heuristic search to outperform leading comparable methods. Together, these contributions systematically improve the practicality and inference efficiency of large language models for long-context tasks.

Email:

Education

2018–2023: Ph.D., Department of Computer Science and Technology, Tsinghua University

2014–2018: B.S., Department of Computer Science and Technology, Tsinghua University

Selected Publications

(\*: Corresponding author; \#: Co-first author)

1. Xiaoang Xu, Shuo Wang*, Xu Han, Zhenghao Liu, Huijia Wu, Peipei Li, Zhiyuan Liu, Maosong Sun, Zhaofeng He*. A*-Thought: Efficient Reasoning via Bidirectional Compression for Low-Resource Settings. NeurIPS. 2025.

2. Zihan Zhou#, Chong Li#, Xinyi Chen#, Shuo Wang*, Yu Chao, Zhili Li, Haoyu Wang, Qi Shi, Zhixing Tan, Xu Han, Xiaodong Shi*, Zhiyuan Liu, Maosong Sun. LLM×Mapreduce: Simplified long-sequence processing using large language models. ACL. 2025.

3. Bowen Ping#, Shuo Wang#, Hanqing Wang, Xu Han, Yuzhuang Xu, Yukun Yan, Yun Chen, Baobao Chang, Zhiyuan Liu*, Maosong Sun*. Delta-CoMe: Training-Free Delta-Compression with Mixed-Precision for Large Language Models. NeurIPS. 2024.

4. Hanqing Wang#, Bowen Ping#, Shuo Wang*, Xu Han, Yun Chen*, Zhiyuan Liu, Maosong Sun. LoRA-Flow: Dynamic LoRA Fusion for Large Language Models in Generative Tasks. ACL. 2024.

5. Shuo Wang, Zhaopeng Tu, Shuming Shi, Yang Liu. On the Inference Calibration of Neural Machine Translation. ACL. 2020.