Diffusion Language Models
Modeling, acceleration, and inference infrastructure.
drew@research ~ $ whoami
Yijie Jin / Drew
Ph.D. @ SJTU Research Intern @ Qwen
I am a Ph.D. student at DENG Lab, Shanghai Jiao Tong University, advised by Prof. Zhijie Deng, and a Research Intern on the Qwen Foundation Model Infrastructure Team.
My research sits at the intersection of Machine Learning Systems and Efficient Generative AI, with a focus on Algorithm–System Co-Design.
research
Algorithm ↔ SystemCo-Design Recursive Self-Improvement
Self-improving model architectures and inference systems.
Efficient Interactive World Models
Algorithm–system co-optimization for real-time efficiency.
selected publications
All papers- ArXiv2026
Wave Forcing: Towards Speed-of-Light Streaming Video Generation
ArXiv Preprint, 2026
ArXivHF Daily #2pdfcodearxiv projectbibtex
@article{jin2026wave, title = {Wave Forcing: Towards Speed-of-Light Streaming Video Generation}, author = {Jin, Yijie and Liu, Yuxuan and Xu, Jiajun and Huang, Ye and Wang, Chuzhi and Li, Jiayang and Chen, Ruiyang and Hong, Chenling and Wang, Xu and Deng, Zhijie}, journal = {ArXiv Preprint}, year = {2026}, } - EMNLP2026
Multi-Block Diffusion Language Models
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2026
MainOralpdfarxivbibtex
@article{jin2026mbd, title = {Multi-Block Diffusion Language Models}, author = {Jin, Yijie and Xu, Jiajun and Liu, Yuxuan and Xu, Chenkai and Tu, Yi and Li, Jiajun and Tu, Dandan and Yan, Xiaohui and Yu, Kai and Liu, Pengfei and Deng, Zhijie}, journal = {Conference on Empirical Methods in Natural Language Processing (EMNLP)}, year = {2026}, } - ICML2026
LightningRL: Breaking the Accuracy–Parallelism Trade-off of Block-wise dLLMs via Reinforcement Learning
International Conference on Machine Learning (ICML), 2026
Regularpdfcodearxivbibtex
@article{hu2026lightningrl, title = {LightningRL: Breaking the Accuracy–Parallelism Trade-off of Block-wise dLLMs via Reinforcement Learning}, author = {Hu, Yanzhe and Jin, Yijie and Liu, Pengfei and Yu, Kai and Deng, Zhijie}, journal = {International Conference on Machine Learning (ICML)}, year = {2026}, } - ArXiv2025
LoPA: Scaling dLLM Inference via Lookahead Parallel Decoding
ArXiv Preprint, 2025
ArXivpdfcodearxivbibtex
@article{xu2025lopa, title = {LoPA: Scaling dLLM Inference via Lookahead Parallel Decoding}, author = {Xu, Chenkai and Jin, Yijie and Li, Jiajun and Tu, Yi and Long, Guoping and Tu, Dandan and Song, Mingcong and Si, Hongjie and Hou, Tianqi and Yan, Junchi and Deng, Zhijie}, journal = {ArXiv Preprint}, year = {2025}, } - ICLR2026
Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing
International Conference on Learning Representations (ICLR), 2026
Posterpdfcodearxivbibtex
@article{wang2025d2f, title = {Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing}, author = {Wang, Xu and Xu, Chenkai and Jin, Yijie and Jin, Jiachun and Zhang, Hao and Deng, Zhijie}, journal = {International Conference on Learning Representations (ICLR)}, year = {2026}, } - ACL2025
Multimodal Transformers are Hierarchical Modal-wise Heterogeneous Graphs
Annual Meeting of the Association for Computational Linguistics (ACL), 2025
MainPosterpdfcodebibtex
@article{jin2025gsit, title = {Multimodal Transformers are Hierarchical Modal-wise Heterogeneous Graphs}, author = {Jin, Yijie and Peng, Junjie and Lin, Xuanchao and Yuan, Haochen and Wang, Lan and Zheng, Cangzhi}, journal = {Annual Meeting of the Association for Computational Linguistics (ACL)}, year = {2025}, }
news
Archive-
Our new preprint Wave Forcing reached #2 on Hugging Face Daily Papers. Project · Code.
-
MBD-LMs selected for an Oral presentation at EMNLP 2026 (Main).
-
MBD-LMs accepted at EMNLP 2026 (Main).
-
LightningRL accepted at ICML 2026 (Regular).
-
Received the Shanghai Outstanding Graduate Award.
-
D2F accepted at ICLR 2026 (Poster).
awards
- Jul, 2026Shanghai Outstanding Graduate Award
- May, 2025ASC Student Supercomputer Challenge First Prize(Preliminary Round 5th, Final Round 7th, AlphaFold3 & DeepSeek-R1 Inference Optimization).
- Nov, 2024Shanghai University Academic Excellence First Class Scholarship
- Aug, 2024Asia and Pacific Mathematical Contest in Modeling First Prize
- Apr, 2024ASC Student Supercomputer Challenge Finalist(Core Task: LLM Inference Optimization).
- Nov, 2023Shanghai University Academic Excellence Top Class Scholarship
contact
Feel free to reach out via email.