Double-click to toggle motion / 双击切换动效

Dawei Zhu

Hi! I am a Ph.D. student at the School of Computer Science, Peking University, advised by Prof. Sujian Li.

My research focuses on long-context language modeling and retrieval. I am also interested in agents, alignment, multimodal models, and AI research tools. Before that, I received my bachelor's degree from the School of EECS, Peking University.

Email Google Scholar Github Semantic Scholar
Dawei Zhu research interests
Double-click to toggle motion / 双击切换动效
HOVER IMAGES TO SHOW ABSTRACTS

Representative Research Works

PaperBanana PaperBanana
PaperBanana: Automating Academic Illustration for AI Scientists
Dawei Zhu, Rui Meng, Yale Song, Xiyu Wei, Sujian Li, Tomas Pfister, Jinsung Yoon
arXiv, 2026
An agentic framework for generating publication-ready academic diagrams and plots from research text.
arXiv Project Code
DocLens DocLens
DocLens: A Tool-Augmented Multi-Agent Framework for Long Visual Document Understanding
Dawei Zhu, Rui Meng, Jiefeng Chen, Sujian Li, Tomas Pfister, Jinsung Yoon
arXiv, 2025
Multi-agent visual document understanding with evidence localization and answer adjudication.
arXiv
LongRePS LongRePS
Chain-of-Thought Matters: Improving Long-Context LMs with Reasoning Path Supervision
Dawei Zhu*, Xiyu Wei*, Guangxiang Zhao, Wenhao Wu, Haosheng Zou, et al.
Findings of EMNLP, 2025
LongRePS trains long-context models to generate reliable reasoning paths for stronger long-context QA.
Paper arXiv
LongEmbed LongEmbed
LongEmbed: Extending Embedding Models for Long Context Retrieval
Dawei Zhu, Liang Wang, Nan Yang, Yifan Song, Wenhao Wu, Furu Wei, Sujian Li
EMNLP, 2024
A benchmark and training recipe for long-context retrieval, extending embedding models up to 32K tokens.
Paper Code
SELECTED AND RECENT

Publications

Long Context & Retrieval

PaperBanana PaperBanana
PaperBanana: Automating Academic Illustration for AI Scientists
Dawei Zhu, Rui Meng, Yale Song, Xiyu Wei, Sujian Li, Tomas Pfister, Jinsung Yoon
arXiv, 2026
arXiv / Project / Code
DocLens DocLens
DocLens: A Tool-Augmented Multi-Agent Framework for Long Visual Document Understanding
Dawei Zhu, Rui Meng, Jiefeng Chen, Sujian Li, Tomas Pfister, Jinsung Yoon
arXiv, 2025
arXiv
LongRePS LongRePS
Chain-of-Thought Matters: Improving Long-Context Language Models with Reasoning Path Supervision
Dawei Zhu*, Xiyu Wei*, Guangxiang Zhao, Wenhao Wu, Haosheng Zou, Junfeng Ran, Xun Wang, Lin Sun, Xiangzheng Zhang, Sujian Li
Findings of EMNLP, 2025
Paper / arXiv
LongEmbed LongEmbed
LongEmbed: Extending Embedding Models for Long Context Retrieval
Dawei Zhu, Liang Wang, Nan Yang, Yifan Song, Wenhao Wu, Furu Wei, Sujian Li
EMNLP, 2024
Paper / Code / Data
PoSE PoSE
PoSE: Efficient Context Window Extension of LLMs via Positional Skip-wise Training
Dawei Zhu, Nan Yang, Liang Wang, Yifan Song, Wenhao Wu, Furu Wei, Sujian Li
ICLR, 2024
Paper / Code
LongAttn LongAttn
LongAttn: Selecting Long-context Training Data via Token-level Attention
Longyun Wu, Dawei Zhu, Guangxiang Zhao, Zhuocheng Yu, Junfeng Ran, Xiangyu Wong, Lin Sun, Sujian Li
Findings of ACL, 2025
Paper / arXiv

Agents, Evaluation & Multimodal

MiMo MiMo
MiMo-VL Technical Report
Xiaomi LLM-Core Team, including Dawei Zhu
arXiv, 2025
arXiv
MiMo MiMo
MiMo: Unlocking the Reasoning Potential of Language Model -- From Pretraining to Posttraining
LLM-Core Xiaomi, including Dawei Zhu
arXiv, 2025
arXiv
CoUDA CoUDA
CoUDA: Coherence Evaluation via Unified Data Augmentation
Dawei Zhu*, Wenhao Wu, Yifan Song, Fangwei Zhu, Ziqiang Cao, Sujian Li
NAACL, 2024
Paper / Code
RestGPT RestGPT
RestGPT: Connecting Large Language Models with Real-World Applications via RESTful APIs
Yifan Song, Weimin Xiong, Dawei Zhu, Cheng Li, Ke Wang, Ye Tian, Sujian Li
arXiv, 2023
arXiv / Project
FairEval FairEval
Large Language Models are not Fair Evaluators
Peiyi Wang, Lei Li, Liang Chen, Dawei Zhu, Binghuai Lin, Yunbo Cao, Qi Liu, Tianyu Liu, Zhifang Sui
arXiv, 2023
arXiv / Code

* denotes equal contribution. For a full list, please see Google Scholar.

EDUCATION & EXPERIENCE

Education & Work Experience

Peking University Beijing, China
Ph.D. Student, School of Computer Science Sept 2022 - Present
Advisor: Prof. Sujian Li.
B.Sc. in Computer Science and Technology, School of EECS Sept 2018 - Jun 2022
Microsoft Research Asia Beijing, China
Research Internship Jun 2023 - Mar 2024
Mentored by Liang Wang and Nan Yang. Research on long-context retrieval and efficient context window extension.
AWARDS & COMMUNITY

Awards & Community Contributions