Yingjia Wan

Yingjia Wan

Student Researcher in Natural Language Processing

About

Hi! I am Yingjia, currently a student researcher in NLP. I obtained my master’s degree at the Language Technology lab, University of Cambridge. My research interests include model reasoning, generation evaluation, fact-checking, and model alignment, etc. I received my bachelor’s degree from University of Macau where I graduated top of the cohort as a valedictorian.

- Collaboration: I feel incredibly grateful to work with some outstanding NLP researchers and mentors such as: Dr. Zhijiang Guo, Dr. Ivan Vulić, Jianqiao Lu, and Dr. Yinya Huang. The list goes on!

- Services: Reviewer for ACL 2024, EMNLP 2024, ICLR 2025, and NAACL 2025.

- Status: I am applying for PhD admission 2025 in NLP and always open for collaboration. Feel free to drop me an email for a chat!

Interests
  • LLM Reasoning
  • Generation Evaluation
  • Automated Fact-Checking
  • AI4Math
  • LLM Alignment
Education

  • M.Phil. in Linguistics (NLP track), 2022-2023

    University of Cambridge

  • Funded Exchange, 2019-2020

    University of Michigan

  • B.A. in English Studies, 2017-2021

    University of Macau

Recent Publications

Jianqiao Lu, Yingjia Wan, Yinya Huang, Zhengying Liu, Zhijiang Guo (2024). FormalAlign: Automated Alignment Evaluation for Autoformalization. Preprint available.

arXiv

Zhongshen Zeng, Yinhong Liu, Yingjia Wan, Jingyao Li, Pengguang Chen, Jianbo Dai, Yuxuan Yao, Rongwu Xu, Zehan Qi, Wanru Zhao, Linling Shen, Jianqiao Lu, Haochen Tan, Yukang Chen, Hao Zhang, Zhan Shi, Bailin Wang, Zhijiang Guo, Jiaya Jia (2024). MR-Ben: A Meta-Reasoning Benchmark for Evaluating System-2 Thinking in LLMs. NeurIPS 2024.

Code Dataset Blog arXiv

Jianqiao Lu, Yingjia Wan, Zhengying Liu, Yinya Huang, Haiming Wang, Zhicheng Yang, Jing Tang, Zhijiang Guo (2024). Process-Driven Autoformalization in Lean 4. Preprint available.

Code Dataset arxiv

Jianqiao Lu, Zhiyang Dou, Hongru Wang, Zeyu Cao, Jianbo Dai, Yingjia Wan, Yinya Huang, Zhijiang Guo (2024). AutoPSV: Automated Process-Supervised Verifier. NeurIPS 2024.

Code arxiv

Yingjia Wan, Ivan Vulić (2023). Multimodal Prompt Tuning for Cognition-Enhanced NLP. [MPhil Thesis].

PDF

See all publications

Research Experience

 
 
 
 
 
Uniersity of Cambridge
Student Researcher (remote)
Uniersity of Cambridge
March 2024 – Present Cambridge, UK
 
 
 
 
 
Hong Kong University of Science and Technology - Guangzhou
Research Assistant (remote)
Hong Kong University of Science and Technology - Guangzhou
March 2024 – August 2024
Topic: LLMs for Formal Mathematics, Automated Thoerem Proving, LLM Reasoning.
 
 
 
 
 
Centre for Cognitive and Brain Sciences, Uniersity of Macau
Research Assistant
Centre for Cognitive and Brain Sciences, Uniersity of Macau
January 2021 – February 2022 Macao S.A.R, China
Topic: Bilinguals’ Word-Level Attention During Listening Assessments.
 
 
 
 
 
Depts of Computer Science & of Psychology, University of Macau
Research Intern
Depts of Computer Science & of Psychology, University of Macau
July 2019 – December 2020 Macao S.A.R, China
Topic: Applying NLP in Developmental Psychology.
 
 
 
 
 
Department of Education & of Psychology, University of Michigan
Lab Member (hybrid)
Department of Education & of Psychology, University of Michigan
April 2020 – July 2021 Ann Arbor, MI, USA
Topic: A Cross-Cultural Perpective Survey at Universities in USA and China during the Pandemic.

Accomplish­ments

BlueDot Impact and University of Cambridge
Finished the excellent learning+project course on AI safety and alignment. Shoutout to the org team, my facilitators and the June 2024 cohort!
Macau Foundation
Fundação Macau Academic Prize
Macau Foundation Jul 2021
University of Macau
Spoke at the graduation ceremony representing the undergraduate class of 2021.
University of British Columbia
The beginning of my NLP journey!
Honours College at University of Macau
Full waiver scholarship covering 4-year tuition, college accommodation, and semester exchange.

Blogs

Regulations and Explorations in AI Safety Control - Efforts, Challenges, & Future Directions
The topic of AI’s existential crisis for humanity has been sparked by the recent development of AI technologies with their transformative power from chatbots, autonomous vehicles to advanced healthcare diagnostics. The seamless integration of AI technology with the untransparent nature of its internal agorithms based on deep learning brings forth significant concerns from the academia, industry, government, and general public about its safety and robustness.
Last updated on July 3, 2023

Contact