Chen Zhang (张晨)
Ph.D. Student @ Peking University
I am a final-year Ph.D. student at PIE Lab, Wangxuan Institute of Computer Technology, Peking University, advised by Prof. Yansong Feng.
My research aims to make LLMs more inclusive and intelligent across diverse human languages.
I am currently seeking postdoctoral opportunities. If you are interested in my profile, please feel
free to reach out!
- NLP for Low-Resource Languages: Enhancing the transparency, inclusivity, and efficiency in language technology for underrepresented languages.
- LLMs for Language Science: Leveraging LLMs to analyze linguistic phenomena, such as language contact and acquisition, for insights into how human languages function and evolve.
- Complex Reasoning with LLMs: Unlocking the potential of LLMs to tackle complex problems and building trustworthy, socially beneficial systems for specialized domains.
- [NOV 2025] Involved in organizing the WiNLP 2025 workshop at EMNLP 2025.
- [MAY 2025] Our papers on low-resource languages are accepted to ACL 2025.
- [MAR 2025] Released MiLiC-Eval, an NLP evaluation suite for China's minority languages.
No. 128 Zhongguancun North Street
Haidian District, Beijing, 100871
zhangch [at] pku [dot] edu [dot] cn
* denotes equal contribution.
Following the #BenderRule, languages are specified for each work.
Read it in Two Steps: Translating Extremely Low-Resource Languages with
Code-Augmented Grammar Books
ACL 2025
*, Jiuheng Lin*, Xiao Liu, Zekai Zhang, Yansong Feng
[paper] [github]
Zhuang, Kalamang
*, Jiuheng Lin*, Xiao Liu, Zekai Zhang, Yansong Feng
[paper] [github]
Zhuang, Kalamang
Cross-Lingual Transfer of Cultural Knowledge: An Asymmetric Phenomenon
ACL 2025
, Zhiyuan Liao, Yansong Feng
[paper] [github]
English, Chinese, Korean, Tibetan, Mongolian
, Zhiyuan Liao, Yansong Feng
[paper] [github]
English, Chinese, Korean, Tibetan, Mongolian
MiLiC-Eval: Benchmarking Multilingual LLMs for China's Minority Languages
ACL 2025 (Findings)
, Mingxu Tao, Zhiyuan Liao, Yansong Feng
[paper] [github] [huggingface]
Tibetan, Uyghur, Kazakh, Mongolian
, Mingxu Tao, Zhiyuan Liao, Yansong Feng
[paper] [github] [huggingface]
Tibetan, Uyghur, Kazakh, Mongolian
Eliciting and Improving the Causal Reasoning Abilities of Large Language Models
with Conditional Statements
Computational Linguistics 2025
Xiao Liu, Da Yin, , Yansong Feng, Dongyan Zhao
[paper]
English
2024
Xiao Liu, Da Yin, , Yansong Feng, Dongyan Zhao
[paper]
English
Unlocking the Potential of Model Merging for Low-Resource Languages
EMNLP 2024 (Findings)
Mingxu Tao*, *, Quzhe Huang*, Tianyao Ma, Songfang Huang, Dongyan Zhao, Yansong Feng
[paper] [huggingface]
Tibetan, Uyghur, Mongolian, Tamil, Telugu, Odia, Bengali
Mingxu Tao*, *, Quzhe Huang*, Tianyao Ma, Songfang Huang, Dongyan Zhao, Yansong Feng
[paper] [huggingface]
Tibetan, Uyghur, Mongolian, Tamil, Telugu, Odia, Bengali
Teaching Large Language Models an Unseen Language on the Fly
ACL 2024 (Findings)
, Xiao Liu, Jiuheng Lin, Yansong Feng
[paper] [github] [website]
Zhuang, Kalamang, and other 7 mid-resource languages
, Xiao Liu, Jiuheng Lin, Yansong Feng
[paper] [github] [website]
Zhuang, Kalamang, and other 7 mid-resource languages
MC2: Towards Transparent and Culturally-Aware NLP for Minority
Languages in China
ACL 2024
*, Mingxu Tao*, Quzhe Huang*, Jiuheng Lin*, Zhibin Chen, Yansong Feng
[paper] [github] [website]
Tibetan, Uyghur, Kazakh, Mongolian
*, Mingxu Tao*, Quzhe Huang*, Jiuheng Lin*, Zhibin Chen, Yansong Feng
[paper] [github] [website]
Tibetan, Uyghur, Kazakh, Mongolian
Harder Task Needs More Experts: Dynamic Routing in MoE Models
ACL 2024
Quzhe Huang*, Zhenwei An*, Nan Zhuang, Mingxu Tao, , Yang Jin, Kun Xu, Kun Xu, Liwei Chen, Songfang Huang, Yansong Feng
[paper]
English
Quzhe Huang*, Zhenwei An*, Nan Zhuang, Mingxu Tao, , Yang Jin, Kun Xu, Kun Xu, Liwei Chen, Songfang Huang, Yansong Feng
[paper]
English
Can LLMs Learn a New Language on the Fly? A Case Study on Zhuang
ICLR 2024 Tiny Paper
, Mingxu Tao, Quzhe Huang, Zhibin Chen, Yansong Feng
[paper]
Zhuang
, Mingxu Tao, Quzhe Huang, Zhibin Chen, Yansong Feng
[paper]
Zhuang
Can Perplexity Reflect Large Language Model's Ability in Long Text
Understanding?
ICLR 2024 Tiny Paper
Yutong Hu, Quzhe Huang, Mingxu Tao, , Yansong Feng
[paper]
English
2023
Yutong Hu, Quzhe Huang, Mingxu Tao, , Yansong Feng
[paper]
English
Lawyer LLaMA: Enhancing LLMs with Legal Knowledge
arXiv 2305.15062
Quzhe Huang*, Mingxu Tao*, *, Zhenwei An*, Cong Jiang, Zhibin Chen, Zirui Wu, Yansong Feng
[preprint] [github]
Chinese
Quzhe Huang*, Mingxu Tao*, *, Zhenwei An*, Cong Jiang, Zhibin Chen, Zirui Wu, Yansong Feng
[preprint] [github]
Chinese
How Many Answers Should I Give? An Empirical Study of Multi-Answer Reading
Comprehension
ACL 2023 (Findings)
, Jiuheng Lin, Xiao Liu, Yuxuan Lai, Yansong Feng, Dongyan Zhao
[paper] [github]
English
, Jiuheng Lin, Xiao Liu, Yuxuan Lai, Yansong Feng, Dongyan Zhao
[paper] [github]
English
The Magic of IF: Investigating Causal Reasoning Abilities in Large Language
Models of Code
ACL 2023 (Findings)
Xiao Liu, Da Yin, , Yansong Feng, Dongyan Zhao
[paper] [github]
English
Xiao Liu, Da Yin, , Yansong Feng, Dongyan Zhao
[paper] [github]
English
Relation-Aware Question Answering for Heterogeneous Knowledge Graphs
EMNLP 2023 (Findings)
Haowei Du, Quzhe Huang, Chen Li, , Yang Li, Dongyan Zhao
[paper]
English
Haowei Du, Quzhe Huang, Chen Li, , Yang Li, Dongyan Zhao
[paper]
English
Cross-Lingual Question Answering over Knowledge Base as Reading
Comprehension
EACL 2023 (Findings)
, Yuxuan Lai, Yansong Feng, Xingyu Shen, Haowei Du, Dongyan Zhao
[paper] [github]
Chinese, Persian, German, Romanian, Italian, Russian, French, Dutch, Spanish, Hindi, Portuguese
, Yuxuan Lai, Yansong Feng, Xingyu Shen, Haowei Du, Dongyan Zhao
[paper] [github]
Chinese, Persian, German, Romanian, Italian, Russian, French, Dutch, Spanish, Hindi, Portuguese
UnifEE: Unified Evidence Extraction for Fact Verification
EACL 2023
Nan Hu, Zirui Wu, Yuxuan Lai, , Yansong Feng
[paper] [github]
English
2022 and before
Nan Hu, Zirui Wu, Yuxuan Lai, , Yansong Feng
[paper] [github]
English
Knowledge-Enhanced Iterative Instruction Generation and Reasoning for Knowledge
Base Question Answering
NLPCC 2022
Haowei Du, Quzhe Huang, , Dongyan Zhao
[paper] [preprint] English
Haowei Du, Quzhe Huang, , Dongyan Zhao
[paper] [preprint] English
Extract, Integrate, Compete: Towards Verification Style Reading
Comprehension
EMNLP 2021 (Findings)
, Yuxuan Lai, Yansong Feng, Dongyan Zhao
[paper] [github]
Chinese
, Yuxuan Lai, Yansong Feng, Dongyan Zhao
[paper] [github]
Chinese
A review of deep learning in question answering over knowledge bases
AI Open 2021, Volume 2
, Yuxuan Lai, Yansong Feng, Dongyan Zhao
[paper]
English
, Yuxuan Lai, Yansong Feng, Dongyan Zhao
[paper]
English
- Area Chair: ACL Rolling Review, LREC
- Reviewer: ACL Rolling Review (Great Reviewer x3), ACL, EMNLP, LREC, COLING, NLPCC (Best Reviewer Award 2025)
- Workshop Organizer: Widening NLP (WiNLP) 2025
- Student Volunteer: ACL 2025, ACL 2024, EMNLP 2021 (remote)
- ZhuangBench: A benchmark consisting of a small set of Zhuang–Chinese parallel sentences and a
Zhuang dictionary, designed to evaluate whether LLMs can comprehend an unseen language on the fly.
Now part of LongBench v2.
[github] [paper] - ZhuangRules: A collection of Zhuang grammar rules for evaluating whether LLMs can effectively
leverage grammar books in low-resource language understanding.
[github] [paper] - MC2: A web-crawled corpus covering four minority languages in China:
Tibetan, Uyghur, Kazakh, and Mongolian.
[github] [huggingface] [paper] - MiLiC-Eval: A multi-task evaluation benchmark for four minority languages in China: Tibetan,
Uyghur, Kazakh, and Mongolian.
[github] [huggingface] [paper] - Lawyer LLaMA: A Chinese legal-domain LLM based on Llama 2 trained on synthetic data of legal consultations.
[github] [huggingface] [report]
Teaching Assistant @ Peking University
- Foundations of Natural Language Processing (Spring 2024, 2025)
- Empirical Methods for Natural Language Processing (Spring 2022)
- Data Structures and Algorithms (B) (Fall 2020, Spring 2021)
- President Scholarship (校长奖学金), Peking University (2025)
- Award for Scientific Research (优秀科研奖), Peking University (2024)
- Outstanding Graduate of Beijing (北京市优秀毕业生), Beijing Municipal Education Commission (2021)
- Outstanding Graduate (北京大学优秀毕业生), Peking University (2021)
- Best Project, Google ML Winter Camp (2020) [link]
- Founder Scholarship (方正奖学金), Peking University (2018, 2019)
- My name in Chinese characters is 张晨 (Zhāng Chén, /ʈʂɑŋ˥˥ ʈʂʰən˧˥/). My given name 晨 (Chén) means morning.
- My mother tongue is the Jinsha Dialect (金沙话), a transitional dialect between Mandarin and Wu Chinese.
- Intermediate in Japanese and Spanish; basic knowledge of Korean, German, and Uzbek.
- I enjoy hiking, learning new languages, and exploring craft beers (especially IPAs). If you visit Beijing, I'd be happy to plan a great day hiking in the nearby hills and take you to some excellent local taprooms.