Zejun MA

Principal Researcher, TikTok

Joined

May 2021

Names

Zejun MA (Preferred)

Emails

****@bytedance.com (Confirmed)

Personal Links

Google Scholar

Career & Education History

Principal Researcher

TikTok (bytedance.com)

2024 – Present

Principal Researcher

ByteDance Inc. (bytedance.com)

2017 – 2024

Principal Researcher

iscas (cas.edu)

2007 – 2012

Advisors, Relations & Conflicts

PhD Advisor

Bo XU

2007 – 2012

Expertise

deep learning

speech recognition

natural language processing

reinforcement learning

2006 – Present

Publications

video-SALMONN 2: Caption-Enhanced Audio-Visual Large Language Models
Changli Tang, Yixuan Li, Yudong Yang, Jimin Zhuang, Guangzhi Sun, Wei Li, Zejun MA, Chao Zhang
- Submitted to ICLR 2026
- Readers: Everyone
video-SALMONN S: Streaming Audio-Visual LLMs Beyond Length Limits via Memory
Guangzhi Sun, Yixuan Li, Xiaodong Wu, Yudong Yang, Wei Li, Zejun MA, Chao Zhang
- Submitted to ICLR 2026
- Readers: Everyone
SWE-Perf: Can Language Models Optimize Code Performance on Real-World Repositories?
Xinyi He, Qian Liu, Mingzhe Du, Lin Yan, ZhiJie Fan, Yiming Huang, Zejian Yuan, Zejun MA
- Submitted to ICLR 2026
- Readers: Everyone
First return, entropy-eliciting explore
Tianyu Zheng, Tianshun Xing, Qingshui Gu, Taoran Liang, Xingwei Qu, 周欣, Yizhi LI, Zhoufutu Wen, Chenghua Lin, Wenhao Huang, Qian Liu, Ge Zhang, Zejun MA
- Submitted to ICLR 2026
- Readers: Everyone
SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning
Zhenghai Xue, Longtao Zheng, Qian Liu, Yingru Li, Xiaosen Zheng, Zejun MA, Bo An
- ICLR 2026 Poster
- Readers: Everyone
SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning
Zhenghai Xue, Longtao Zheng, Qian Liu, Yingru Li, Xiaosen Zheng, Zejun MA, Bo An
- MTI-LLM @ NeurIPS 2025 Spotlight
- Readers: Everyone
SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning
Zhenghai Xue, Longtao Zheng, Qian Liu, Yingru Li, Xiaosen Zheng, Zejun MA, Bo An
- DL4C @ NeurIPS 2025 Poster
- Readers: Everyone
SWE-Perf: Can Language Models Optimize Code Performance on Real-World Repositories?
Xinyi He, Qian Liu, Mingzhe Du, Lin Yan, ZhiJie Fan, Yiming Huang, Zejian Yuan, Zejun MA
- DL4C @ NeurIPS 2025 Oral
- Readers: Everyone
LiveCC: Learning Video LLM with Streaming Speech Transcription at Scale
Joya Chen, Ziyun Zeng, Yiqi Lin, Wei Li, Zejun MA, Mike Zheng Shou
- CVPR 2025
- Readers: Everyone
Afterburner: Reinforcement Learning Facilitates Self-Improving Code Efficiency Optimization
Mingzhe Du, Anh Tuan Luu, Yue Liu, Yuhao QING, Dong HUANG, Xinyi He, Qian Liu, Zejun MA, See-Kiong Ng
- ICML 2025 Workshop PRAL
- Readers: Everyone

View all 31 publications

Co-Authors

View all 117 co-authors