Scholar
Yijie Jin
Google Scholar ID: L220uBgAAAAJ
Incoming Ph.D. in Shanghai Jiao Tong University (SJTU)
Citations & Impact
Citations
7
H-index
2
i10-index
0
Publications
2
Co-authors
2
list available
Publications
Resume (English only)
Academic Achievements
- - Publications:
- - "Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing", ArXiv, Authors: Xu Wang, Chenkai Xu, Yijie Jin, Jiachun Jin, Hao Zhang, Zhijie Deng.
- - "Multimodal Transformers are Hierarchical Modal-wise Heterogeneous Graphs", ACL 2025 (Main), Authors: Yijie Jin, Junjie Peng, Xuanchao Lin, Haochen Yuan, Lan Wang, Cangzhi Zheng.
- - Awards:
- - First Prize in the Preliminary Round of ASC 2025 Student Supercomputer Challenge (Worldwide No. 5).
- - NO.7 Worldwide in the Final Round of ASC 2025 Student Supercomputer Challenge (ASC25).
- - Invited Talks:
- - 2025 vLLM Shenzhen Meetup, presented on D2fEngine (the first vLLM implementation for dLLMs in the open-source community).
Research Experience
- - Currently working on D2fEngine, which is the first open-sourced vLLM for dLLMs worldwide.
Education
- - Ph.D. Candidate: Qingyuan Research Institute, School of Computer Science, Shanghai Jiao Tong University, Advisor: Prof. Zhijie Deng.
- - B.Eng. Student: Department of Computer Science and Technology, School of Computer Engineering and Science, Shanghai University, Supervisor: Prof. Junjie Peng.
Background
- - Research Interests: Machine Learning Systems (MLSys), with a strong focus on Efficient Generative AI, traditional NLP, and multimodal learning, particularly in the domain of Multimodal Sentiment Analysis (MSA).
- - Brief Introduction: Incoming Ph.D. student at Qingyuan Research Institute, School of Computer Science, Shanghai Jiao Tong University (SJTU). Currently an Undergrad. student at the Department of Computer Science and Technology, School of Computer Engineering and Science, Shanghai University (SHU).
Miscellany
- - Blogs are posted on Zhihu, covering experience sharing, technical coding tutorials, academic paper discussions, personal career insights, and casual life reflections.
- - Maintaining the MLSys Wiki for self-learning and sharing technical insights.
