Yapeng Mi

I am a phd student of University of Science and Technology of China(USTC) ,advised by Qing Li and Jie Wang. Previously, I receive my bachelor’s degree in Computer Science and Technology from Harbin Institute of Technology(HIT) in 2025.

Currently, I am focusing on research in Multimodal and LLM Agents. Feel free to reach out if you are interested in my work! 🤝

News

Jan 15, 2026	One paper is accepted by ICLR 2026! 👉 MILR
Sep 19, 2025	One paper is accepted by NeurIPS 2025! 👉 SPORT
Apr 10, 2025	New position paper on system-level principles for building LLM agents: See paper for details.

Publications

MILR: Improving Multimodal Image Generation via Test-Time Latent Reasoning

Yapeng Mi , Hengli Li , Yanpeng Zhao^✉ , Chenxi Li , Huimin Wu , Xiaojian Ma , Song-Chun Zhu , Ying Nian Wu , and Qing Li^✉

International Conference on Learning Representations (ICLR), 2026

arXiv Bib Website

@article{2026milr,
  title = {MILR: Improving Multimodal Image Generation via Test-Time Latent Reasoning },
  author = {Mi, Yapeng and Li, Hengli and Zhao, Yanpeng and Li, Chenxi and Wu, Huimin and Ma, Xiaojian and Zhu, Song-Chun and Wu, Ying Nian and Li, Qing},
  journal = {International Conference on Learning Representations (ICLR)},
  year = {2026},
  correspondence = {Zhao, Yanpeng and Li, Qing},
}

Iterative Tool Usage Exploration for Multimodal Agents via Step-wise Preference Tuning

Pengxiang Li* , Zhi Gao* , Bofei Zhang , Yapeng Mi , Xiaojian Ma , Chenrui Shi , Tao Yuan , Yuwei Wu^✉ , Yunde Jia , Song-Chun Zhu , and Qing Li^✉

Advances in Neural Information Processing Systems (NeurIPS), 2025

arXiv Bib Website

@article{2025iterative,
  title = {Iterative Tool Usage Exploration for Multimodal Agents via Step-wise Preference Tuning },
  author = {Li, Pengxiang and Gao, Zhi and Zhang, Bofei and Mi, Yapeng and Ma, Xiaojian and Shi, Chenrui and Yuan, Tao and Wu, Yuwei and Jia, Yunde and Zhu, Song-Chun and Li, Qing},
  equalauthor = { Li, Pengxiang and Gao, Zhi},
  journal = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
  correspondence = {Wu, Yuwei and Li, Qing},
}

Building LLM Agents by Incorporating Insights from Computer Systems

Yapeng Mi , Zhi Gao , Xiaojian Ma , and Qing Li^✉

arXiv preprint arXiv:2504.04485, 2025

arXiv Bib

@article{mi2025building,
  title = {Building LLM Agents by Incorporating Insights from Computer Systems },
  author = {Mi, Yapeng and Gao, Zhi and Ma, Xiaojian and Li, Qing},
  journal = {arXiv preprint arXiv:2504.04485},
  year = {2025},
  correspondence = {Li, Qing},
}