Biography

Guanhua Huang (黄冠华) is an algorithm engineer at Tencent. Previously, he earned his Ph.D. in Data Science from University of Science and Technology of China (USTC), where he was advised by Prof. Weinan E and Prof. Zhouwang Yang. Before that, he received a Bachelor’s degree in Mathematics from University of Science and Technology of China (USTC), advised by Prof. Zhouwang Yang. He completed several internships at BIBDR, ByteDance, and Tencent, where he contributed to fascinating work such as PaSa.

Interests
  • Reasoning LLMs
  • LLM Agents
  • NLP Applications
  • Scholarly Document Processing
Education
  • Ph.D. in Data Science, 2019-2025

    School of Artificial Intelligence and Data Science, University of Science and Technology of China

  • B.S. in Mathematics, 2015-2019

    School of the Gifted Young, University of Science and Technology of China

Experience

 
 
 
 
 
Tencent Hunyuan
Research Intern
Tencent Hunyuan
February 2025 – July 2025 Beijing, China
Worked on Reasoning Model Research.
 
 
 
 
 
ByteDance Research
Research Intern
ByteDance Research
January 2022 – February 2025 Beijing, China
Mentored by Dr. Jiaze Chen and Dr. Yuchen Zhang, working on LLM Research and Information Extraction Research.
 
 
 
 
 
Peking University
Visiting Scholar
Peking University
July 2021 – July 2025 Beijing, China
Advised by Prof. Weinan E, working on LLM Research, Information Extraction Research and AI4Science Research.
 
 
 
 
 
Beijing Institute of Big Data Research
Intern
Beijing Institute of Big Data Research
January 2019 – December 2021 Beijing, China
Advised by Dr. Yi Zhang, working on research and projects of Big Data in Economics.
 
 
 
 
 
Peking University
Visiting Scholar
Peking University
January 2019 – January 2020 Beijing, China
Advised by Prof. Weinan E and Prof. Yucheng Yang, working on Economic Data Science Research.

Recent Publications

Quickly discover relevant content by filtering publications.
(2025). PaSa: An LLM Agent for Comprehensive Academic Paper Search. In The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025).

PDF Code Dataset Project Video

(2024). Improving Generalization and Convergence by Enhancing Implicit Regularization. In Neural Information Processing Systems 37 (NeurIPS 2024).

PDF Code

(2024). AGILE: A Novel Framework of LLM Agents. In Neural Information Processing Systems 37 (NeurIPS 2024).

PDF Code

(2024). Are AI-Generated Text Detectors Robust to Adversarial Perturbations?. In The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024).

PDF Code

(2024). Towards document-level event extraction via Binary Contrastive Generation. In Knowledge-Based Systems (KBS 2024).

PDF DOI