Yanggan Gu

Yanggan Gu

PhD Student @ PolyU

Hi, my name is Yanggan Gu (辜仰淦). I am currently a PhD student at Hong Kong Polytechnic University (PolyU), advised by Prof. Hongxia Yang. Before this, I received my Master’s degree from Soochow University in 2025, advised by Prof. Zhenghua Li.

My research trajectory began with syntactic and semantic parsing and has now shifted to Large Language Models (LLMs). My current research focuses on model fusion and trustworthy AI. I have published 8 top-tier conference papers, including ICML, NeurIPS, ACL, EMNLP, SIGIR, and COLING. I also serve as a reviewer for ICML, ICLR, and NeurIPS.

πŸ”₯ News

  • 2026.05: Β πŸŽ‰πŸŽ‰ One paper πŸ“‘ Model Merging Scaling Laws in Large Language Models has been accepted to ICML 2026!
  • 2025.09: Β πŸŽ‰πŸŽ‰ Two papers have been accepted to NeurIPS 2025, including one Spotlight!
  • 2025.02: Β πŸŽ‰πŸŽ‰ Two papers have been accepted to ACL 2025 Findings!

πŸ“ Publications

2026

  • [C8] πŸ“ Yuanyi Wang, Yanggan Gu (Co-1st), Yiming Zhang, Qi Zhou, Zhaoyi Yan, Congkai Xie, Xinyao Wang, Jianbo Yuan, Hongxia Yang. Model Merging Scaling Laws in Large Language Models. In ICML 2026. πŸ”—[Paper] 🌐[Project] πŸ’»[Code]

  • [J2] πŸ“š Qi Zhou, Yiming Zhang, Yanggan Gu, Yuanyi Wang, Zhaoyi Yan, Zhen Li, Chi Yung Chung, Hongxia Yang. Model Fusion for Scalable and Sustainable Artificial Intelligence: A Review and Outlook. In Journal of Modern Power Systems and Clean Energy, 2026. πŸ”—[Paper]

  • [Pre1] πŸ§ͺ Yuanyi Wang, Yanggan Gu, Zihao Wang, Kunxi Li, Yifan Yang, Zhaoyi Yan, Congkai Xie, Jianmin Wu, Hongxia Yang. MergePipe: A Budget-Aware Parameter Management System for Scalable LLM Merging. In arXiv 2026. πŸ”—[Paper] πŸ’»[Code]

  • [Pre2] πŸ§ͺ Yuanyi Wang, Yifan Yang, Su Lu, Yanggan Gu, Pengkai Wang, Wenjun Wang, Zhaoyi Yan, Congkai Xie, Jianmin Wu, Jialun Cao, Shing-Chi Cheung, Hongxia Yang. Geometry Conflict: Explaining and Controlling Forgetting in LLM Continual Post-Training. In arXiv 2026. πŸ”—[Paper] πŸ’»[Code]

  • [Pre3] πŸ§ͺ Yanggan Gu, Shuo Cai, Zihao Wang, Wenjun Wang, Yuanyi Wang, Pengkai Wang, Sirui Huang, Su Lu, Jianmin Wu, Hongxia Yang. FeatCal: Feature Calibration for Post-Merging Models. In arXiv 2026. πŸ”—[Paper] πŸ’»[Code]

  • [Pre4] πŸ§ͺ Wenjun Wang, Yanggan Gu (Co-1st), Shuo Cai, Yuanyi Wang, Pengkai Wang, Jianmin Wu, Hongxia Yang. E-PMQ: Expert-Guided Post-Merge Quantization with Merged-Weight Anchoring. In arXiv 2026. πŸ”—[Paper] πŸ’»[Code]

2025

  • [C6] πŸ“ Yanggan Gu, Yuanyi Wang, Zhaoyi Yan, Yiming Zhang, Qi Zhou, Fei Wu, Hongxia Yang. InfiFPO: Implicit Model Fusion via Preference Optimization in Large Language Models. In NeurIPS 2025 Spotlight. πŸ”—[Paper] πŸ’»[Code] πŸ€—[Model]

  • [C7] πŸ“ Yuanyi Wang, Zhaoyi Yan, Yiming Zhang, Qi Zhou, Yanggan Gu, Fei Wu, Hongxia Yang. InfiGFusion: Graph-on-Logits Distillation via Efficient Gromov-Wasserstein for Model Fusion. In NeurIPS 2025. πŸ”—[Paper] 🌐[Project] πŸ’»[Code] πŸ€—[Model]

  • [J1] πŸ“š Qi Zhou, Yiming Zhang, Yanggan Gu, Yuanyi Wang, Zhijie Sang, Zhaoyi Yan, Zhen Li, Shengyu Zhang, Fei Wu, Hongxia Yang. Democratizing AI Through Model Fusion: A Comprehensive Review and Future Directions. In Nexus. πŸ”—[Paper]

  • [C5] πŸ“ Yunkai Dang, Mengxi Gao, Yibo Yan, Xin Zou, Yanggan Gu, Aiwei Liu, Xuming Hu. Exploring Response Uncertainty in MLLMs: An Empirical Evaluation under Misleading Scenarios. In EMNLP 2025 Main. πŸ”—[Paper]

  • [C2] πŸ“ Yanggan Gu, Junzhuo Li, Sirui Huang, Xin Zou, Zhenghua Li, Xuming Hu. Capturing Nuanced Preferences: Preference-Aligned Distillation for Small Language Models. In ACL 2025 Findings. πŸ”—[Paper]

  • [C3] πŸ“ Sirui Huang, Yanggan Gu (Co-1st), Zhonghao Li, Xuming Hu, Li Qing, Guandong Xu. StructFact: Reasoning Factual Knowledge from Structured Data with Large Language Models. In ACL 2025 Findings. πŸ”—[Paper]

  • [C4] πŸ“ Sirui Huang, Hanqian Li, Yanggan Gu, Xuming Hu, Qing Li, Guandong Xu. HyperG: Hypergraph-Enhanced LLMs for Structured Knowledge. In SIGIR 2025. πŸ”—[Paper]

2024

  • [C1] πŸ“ Yanggan Gu, Yang Hou, Zhefeng Wang, Xinyu Duan, Zhenghua Li. High-Order Joint Constituency and Dependency Parsing. In COLING 2024. πŸ”—[Paper]

πŸŽ– Honors and Awards

  • πŸ₯‡ 2023.08: 1st place in the CCL-2023 Chinese AMR Parsing Evaluation Task. πŸ”—Report
  • πŸ… 2021.12: Top 10 Innovative Figures of Guangdong University of Technology. πŸͺ§Sohu News
  • πŸ… 2021.11: National Encouragement Scholarship. πŸ”—Link
  • πŸ₯‡ 2021.11: 1st Prize at the 9th β€œChina Software Cup” College Student Design Competition. πŸ”—Link

πŸͺ΄ Service

  • Reviewer: ICML 2026, ICLR 2026, NeurIPS 2026.

πŸŽ“ Education

  • 2025.09 - Present, Ph.D. Student, The Hong Kong Polytechnic University.
  • 2022.09 - 2025.06, M.S., Soochow University.
  • 2018.09 - 2022.06, B.S., Guangdong University of Technology.

πŸ’» Internships

  • 2025.06 - 2025.12, InfiX-ai, Shenzhen, China.