Xiangchi Yuan

Hi, I am Xiangchi Yuan (pronounced as Shyang-Chee Yoo-an/袁象驰), a PhD student at the School of Computer Science at Georgia Institute of Technology🐝. I'm fortunate to work with Prof. Wenke Lee.

I earned my MSCS degree at Brandeis University and earned my bachelor degrees from UESTC and University of Glasgow.

I sincerely thank all the co-authors, especially my mentor 🌲Mr. Chunhui Zhang who generously helped me start my research.

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn

profile photo
News

  • [Jan 2026] I will be joining ByteDance Seed as an Incoming Research Intern.
  • [Aug 2025] One paper accepted to EMNLP'25.
  • [May 2025] I joined Research Intern at Adobe, working on Reasoning Post-training.
  • [May 2025] Two papers accepted to ACL'25 and one paper accepted to ICML'25.
Papers

My research interests include training large foundation models to reason, solving complex tasks, especially for agentic long-context, and security tasks

I appreciate the brilliance and dedication of my co-authors in these works!!

Behavior Knowledge Merge in Reinforced Agentic Models
Xiangchi Yuan, Dachuan Shi, Chunhui Zhang, Zheyuan Liu, Shenglong Yao, Soroush Vosoughi, Wenke Lee
Preprint, 2025.
project page / preprint

Mitigating Forgetting Between Supervised and Reinforcement Learning Yields Stronger Reasoners
Xiangchi Yuan, Xiang Chen, Tong Yu, Dachuan Shi, Can Jin, Wenke Lee, Saayan Mitra
Preprint, 2025.
project page / preprint

SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs
Dachuan Shi, Abedelkadir Asi, Keying Li, Xiangchi Yuan, Leyan Pan, Wenke Lee, Wen Xiao
Preprint, 2025.
project page / preprint

Superficial Self-Improved Reasoners Benefit from Model Merging
Xiangchi Yuan, Chunhui Zhang, Zheyuan Liu, Dachuan Shi, Leyan Pan, Soroush Vosoughi, Wenke Lee
The 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2025.
project page / preprint

LaCache: Ladder-Shaped KV Caching for Efficient Long-Range Generation of Large Language Models
Dachuan Shi, Yonggan Fu, Xiangchi Yuan, Zhongzhi Yu, Haoran You, Sixu Li, Xin Dong, Jan Kautz, Pavlo Molchanov, Yingyan Celine Lin
International Conference on Machine Learning (ICML), 2025.
project page / preprint

Modality-Aware Neuron Pruning for Unlearning in Multimodal Large Language Models
Zheyuan Liu, Guangyao Dou, Xiangchi Yuan, Chunhui Zhang, Zhaoxuan Tan, Meng Jiang
The 63rd Annual Meeting of the Association for Computational Linguistics (ACL), 2025.
project page / preprint

Growing Through Experience: Scaling Episodic Grounding in Language Models Enlargement
Chunhui Zhang, Sirui Wang, Zhongyu Ouyang, Xiangchi Yuan, Soroush Vosoughi
The 63rd Annual Meeting of the Association for Computational Linguistics (ACL Oral Presentation), 2025.
project page / preprint

LongMamba: Enhancing Mamba’s Long-Context Capabilities via Training-Free Receptive Field Enlargement
Zhifan Ye, Kejing Xia, Yonggan Fu, Xin Dong, Jihoon Hong, Xiangchi Yuan, Shizhe Diao, Jan Kautz, Pavlo Molchanov, Yingyan Celine Lin
International Conference on Learning Representations (ICLR), 2025.
project page / preprint

AmoebaLLM: Constructing Any-Shape Large Language Models for Efficient and Instant Deployment
Yonggan Fu, Zhongzhi Yu, Junwei Li, Jiayi Qian, Yongan Zhang, Dachuan Shi, Xiangchi Yuan, Roman Yakunin, Yingyan Celine Lin.
Neural Information Processing Systems (NeurIPS 2024).
project page / preprint

Graph Cross Supervised Learning via Generalized Knowledge
Xiangchi Yuan, Yijun Tian, Chunhui Zhang, Yanfang Ye, Nitesh V Chawla, Chuxu Zhang
ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD 2024).
project page / preprint

Mitigating Severe Robustness Degradation on Graphs
Xiangchi Yuan*, Chunhui Zhang*, Yijun Tian, Yanfang Ye, Chuxu Zhang
International Conference on Learning Representations (ICLR 2024).
project page / preprint

An End-to-End 12-Leading Electrocardiogram Diagnosis System Based on Deformable Convolutional Neural Network With Good Antinoise Ability.
Lang Qin, Yuntao Xie, Xinwen Liu, Xiangchi Yuan, and Huan Wang.
IEEE Transactions on Instrumentation and Measurement (IEEE TIM 2021). .
project page / preprint

To all students/friends

My role model 🌲Mr. Chunhui Zhang gave me tons of precious support and guidance from research to life. I hope to follow his steps to spread kindness like him :)

Feel free to drop me an email if you're up for a laid-back chat about life, career, or research. I'm dedicating (at least) 30 minutes each week for these discussions, and I'm particularly eager to connect with students from underrepresented backgrounds or those dealing with challenges or inequity. Just reach out – I'm here to listen

Fun Stuff

I'm a professional photographer and really enjoy street snaps.

Some Photos!


Design and source code from this cool guy.