👴 About Me

I am currently a first-year Ph.D. student at Fudan University. My research interest mainly includes Computer Vision, AI Agent, Reinforcement Learning, Diffusion. My CV 📄

Email 📧: yangjingyi946@gmail.com or yangjingyi@mail.ustc.edu.cn or jingyiyang25@m.fudan.edu.cn

😀 Please feel free to contact me for communication and collaboration.

🔥 News

📝 Research Topics

Diffusion Large Language Model & Reinforcement Learning

• ArXiv ρ-𝙴𝙾𝚂: Training-free Bidirectional Variable-Length Control for Masked Diffusion LLMs. Jingyi Yang, Yuxian Jiang, Jing Shao^†

• ArXiv DARE: Diffusion Large Language Models Alignment and Reinforcement Executor — an efficient rl training framework for diffusion large language models (based on verl) integrated various dLLM-tailored rl algorithms. Jingyi Yang^{* ‡}, Yuxian Jiang^*, Xuhao Hu^*, Jing Shao^†

• ArXiv Taming Masked Diffusion Language Models via Consistency Trajectory Reinforcement Learning with Fewer Decoding Step. Jingyi Yang, Guanxu Chen, Xuhao Hu, Jing Shao^†

Large Reasoning Model & Reinforcement Learning

• ICLR 2026 Conditional Advantage Estimation for Reinforcement Learning in Large Reasoning Models. Guanxu Chen, Yafu Li, Yuxian Jiang, Chen Qian, Qihan Ren, Jingyi Yang, Yu Cheng, Dongrui Liu, Jing Shao^†

AI Agents & Computer-Use Agent & Self-Evolving Agent

• GitHub WildClawBench: Hard, practical, end-to-end evaluation for AI agents — in the wild. Shuangrui Ding^‡, Xuanlang Dai^*, Long Xing^*, Shengyuan Ding, Ziyu Liu, Jingyi Yang, Penghui Yang, Zhixiong Zhang, Xilin Wei, Yubo Ma, Haodong Duan, Jing Shao, Jiaqi Wang, Dahua Lin, Kai Chen, Yuhang Zang^†

• NeurIPS 2025 RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agents. Jingyi Yang^{* ‡}, Shuai Shao^*, Dongrui Liu, Jing Shao^†

• ICLR 2026 Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents. Shuai Shao, Qihan Ren, Chen Qian, Boyi Wei, Dadi Guo, Jingyi Yang, Xinhao Song, Linfeng Zhang, Weinan Zhang, Dongrui Liu, Jing Shao^†

Video Understanding

• ICLR 2025 Kronecker Mask and Interpretive Prompts are Language-Action Video Learners. Jingyi Yang^*, Zitong Yu^*, Xiuming Ni, Jia He, Hui Li^†

• IEEE TIFS G²V²former: Graph Guided Video Vision Transformer for Face Anti-Spoofing. Jingyi Yang, Zitong Yu^†, Jia He, Xiuming Ni, Liepiao Zhang, Hui Li^†, Xiaochun Cao

Multi-Modal Learning & Domain Generalization for Face Anti-spoofing

• ICCV 2025 DADM: Dual Alignment of Domain Modality for Face Anti-Spoofing. Jingyi Yang^*, Xun Lin^*, Zitong Yu^†, Liepiao Zhang, Xin Liu, Hui Li, Xiaochen Yuan, Xiaochun Cao

• ECAI 2024 Oral Generalized Face Anti-spoofing via Finer Domain Partition and Disentangling Liveness-irrelevant Factors. Jingyi Yang, Zitong Yu, Xiuming Ni, Jia He, Hui Li^†

📖 Selected Publications (* Equal Contribution, ‡ Project Leader, † Corresponding Author)

GitHub

WildClawBench: Hard, practical, end-to-end evaluation for AI agents — in the wild

Shuangrui Ding^‡, Xuanlang Dai^*, Long Xing^*, Shengyuan Ding, Ziyu Liu, Jingyi Yang, Penghui Yang, Zhixiong Zhang, Xilin Wei, Yubo Ma, Haodong Duan, Jing Shao, Jiaqi Wang, Dahua Lin, Kai Chen, Yuhang Zang^†

Project HomePage / Code / Dataset Downloads

Github

DARE: Diffusion Large Language Models Alignment and Reinforcement Executor

Jingyi Yang^*, YuXian Jiang^*, Xuhao Hu^*, Shuang Cheng, Biqing Qi, Jing Shao^†

Code

ArXiv

ρ-𝙴𝙾𝚂: Training-free Bidirectional Variable-Length Control for Masked Diffusion LLMs

Jingyi Yang, Yuxian Jiang, Jing Shao^†

Code

ArXiv

Taming Masked Diffusion Language Models via Consistency Trajectory Reinforcement Learning with Fewer Decoding Step

Jingyi Yang, Guanxu Chen, Xuhao Hu, Jing Shao^†

Code / 机器之心

NeurIPS 2025

RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agents

Jingyi Yang^{* ‡}, Shuai Shao^*, Dongrui Liu, Jing Shao^†

Project HomePage / Code / Dataset Downloads / 机器之心

Demo

ICCV 2025

DADM: Dual Alignment of Domain Modality for Face Anti-Spoofing

Jingyi Yang^*, Xun Lin^*, Zitong Yu^†, Liepiao Zhang, Xin Liu, Hui Li, Xiaochen Yuan, Xiaochun Cao

Code

ICLR 2025

Kronecker Mask and Interpretive Prompts are Language-Action Video Learners

Jingyi Yang^*, Zitong Yu^*, Xiuming Ni, Jia He, Hui Li^†

Code

IEEE TIFS

G²V²former: Graph Guided Video Vision Transformer for Face Anti-Spoofing

Jingyi Yang, Zitong Yu^†, Jia He, Xiuming Ni, Liepiao Zhang, Hui Li^†, Xiaochun Cao

ECAI 2024 Oral 🎉

Generalized Face Anti-spoofing via Finer Domain Partition and Disentangling Liveness-irrelevant Factors

Jingyi Yang, Zitong Yu, Xiuming Ni, Jia He, Hui Li^†

Code

Demo

🎖 Honors and Awards

• 2020, 2021, 2022 Excellent Student Scholarship - First Prize (2%) twice, Third Prize (10%) once
• 2022 Outstanding Graduates of Dalian City
• 2021 Competition Specific Scholarship
• 2021 The Mathematical Contest in Modeling (MCM)/Interdisciplinary Contest in Modeling (ICM) Honorable Mention
• 2019 The 11th National College Student Mathematics Competition Liaoning Province - Third Prize
• 2019 The 28th Dalian College Student Mathematics Competition - First Prize

💻 Academic Service

• Reviewer of International Conference on Machine Learning 2026
• Reviewer of International Conference on Learning Representations 2026
• Reviewer of International Conference on Learning Representations 2025
• Reviewer of IEEE International Joint Conference on Biometrics (IJCB) Multimodal Human Behavior Understanding & Generation 2024