Junchi Yao

Junchi Yao

University of Electronic Science and Technology of China (UESTC)

My name is Junchi Yao (pronounced “JOON-chee YOW”). I am currently in my final year of my undergraduate program in Information System and Information Management from University of Electronic Science and Technology of China (UESTC). Currently, I’m a visiting student in the Machine Learning Department at Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), supervised by Prof. Lijie Hu. I was a research intern at Shanghai AI Lab, where I have the privilege of working with Researcher Peng Ye. Before that, I gained valuable research experience as a research intern at King Abdullah University of Science and Technology (KAUST) under the guidance of Prof. Di Wang.

My research focuses on Large Language Models, particularly in explainability (XAI), LLM agents, and LLM4Science, including social science and physics. Moreover, I have recently begun learning about Vision-Language-Action (VLA) models. My goal is to advance robotics and LLM research toward interpretable, robust, and impactful real-world applications.

I am actively seeking research collaborators, whether you are new or experienced.. Feel free to reach out, or learn more from My CV.

AI Researcher

  • Research focus on LLMs and VLAs
  • Visit at MBZUAI
  • Intern at Shanghai AI Lab
  • Publications in ACL, NeurIPS

World Explorer

  • Visited 12 countries worldwide
  • Traveled to 28 provinces and territories in China
  • Rich experience in driving and hiking

News

  • 2026.01:  🎉 1 Paper is accepted by The Fourteenth International Conference on Learning Representations (ICLR 2026).
  • 2025.11:  ⭐ P1 Technical Report is published on Hugging Face/Arxiv, ranked #2 in Daily Papers, and was featured by QbitAI (量子位).
  • 2025.11:  ✈️ I have joined MBZUAI as a Visiting Student, supervised by Lijie Hu, where I focus on XAI on (M)LLM/ALM/VLA.
  • 2025.09:  🎉 1 Paper is accepted by The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS 2025). See you in San Diago, America!
  • 2025.06:  🎉 1 Paper is accepted by The 42nd International Conference on Machine Learning（ICML 2025）Multi-Agent System Workshop.
  • 2025.05:  🎉 2 Papers are accepted by The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025) Findings. See you in Vienna, Austria!
  • 2025.04:  🎉 1 Paper is accepted by Nature - Scientific Reports.
  • 2025.03:  ✈️ I have joined Shanghai AI Lab as a Research Intern under the guidance of Researcher Peng Ye, where I focus on LLM Agents and LLM for Physics.

Publications

ICLR 2026
Pipeline of SCI-Verify

SCI-VERIFIER: SCIENTIFIC VERIFIER WITH THINKING

Shenghe Zheng*, Chenyu Huang*, Fangchen Yu, Junchi Yao, Jingqi Ye, Tao Chen, Ning Ding, Lei Bai, Ganqu Cui, Peng Ye
ICLR 2026
Paper Code
Technical Report
P1-235B-A22B + PhysicsMinions ranks No.1 on IPhO 2025.

P1: Mastering Physics Olympiads with Reinforcement Learning

Jiacheng Chen∗, Qianjia Cheng∗, Fangchen Yu∗, Haiyuan Wan, Yuchen Zhang, Shenghe Zheng, Junchi Yao*, Qingyang Zhang, Haonan He, Yun Luo, Yufeng Zhao, Futing Wang, Li Sheng, Chengxing Xie, Yuxin Zuo, Yizhuo Li, Wenxuan Zeng, Yulun Wu, Rui Huang, Dongzhan Zhou, Kai Chen, Yu Qiao, Lei Bai, Yu Cheng, Ning Ding, Bowen Zhou, Peng Ye, Ganqu Cui
Technical Report
Paper Code
NeurIPS 2025
Pipeline of PHYSICS Dataset

Scaling Physical Reasoning with the PHYSICS Dataset

Shenghe Zheng*, Qianjia Cheng*, Junchi Yao*, Mengsong Wu, Haonan He, Ning Ding, Yu Cheng, Shuyue Hu, Lei Bai, Dongzhan Zhou, Ganqu Cui, Peng Ye
NeurIPS 2025
Paper Code
ICML 2025 Workshop
WandaPlan's Framework

Is Your LLM-Based Multi-Agent a Reliable Real-World Planner? Exploring Fraud Detection in Travel Planning

Junchi Yao*, Jianhua Xu*, Tianyu Xin*, Ziyi Wang, Shenzhe Zhu, Shu Yang, Di Wang
The 42nd International Conference on Machine Learning（ICML 2025）Multi-Agent System Workshop
Paper Code (Incoming)
ACL 2025 Findings
Repeat Curse Framework

Understanding the Repeat Curse in Large Language Models from a Feature Perspective

Junchi Yao*, Shu Yang*, Jianhua Xu, Lijie Hu, Mengdi Li, Di Wang
In The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025) Findings
Paper Code
ACL 2025 Findings
Fraud-R1 Pipeline

Fraud-R1 : A Multi-Round Benchmark for Assessing the Robustness of LLM Against Augmented Fraud and Phishing Inducements

Shu Yang*, Shenzhe Zhu*, Zeyu Wu, Keyu Wang, Junchi Yao, Junchao Wu, Lijie Hu, Mengdi Li, Derek F. Wong, Di Wang
In The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025) Findings
Paper Code
Nature SR
Social Opinion Prediction Framework

Social opinions prediction utilizes fusing dynamics equation with LLM-based agents

Junchi Yao*, Hongjie Zhang*, Jie Ou, Dingyi Zuo, Zheng Yang, Zhicheng Dong
Nature Scientific Reports
ArXiv

Educations

  • 2025.11 - 2026.05:  Visiting Student of Machine Learning Department, MBZUAI.
  • 2023.11 - 2024.04:  Young Future Energy Leader Programme Student, Khalifa University.
  • 2022.09 - 2026.06:  Bachelor of Engineer in Information System and Information Management, UESTC.

Internships

Service

Reviewer

  • 2025: ICML, ICLR