Yingjia Wan

Yingjia Wan

Master’s student in Natural Language Processing (NLP)

Language Technology Lab, University of Cambridge

About

Yingjia Wan (Alisa Wan) is a master’s student in NLP at the University of Cambridge, supervised by Dr Ivan Vulić and Prof Anna Korhonen. Her research interests include multimodal NLP, prompting, debiasing language models, and the intersection of NLP and cognitive science. Her master’s dissertation was on multimodal prompting with human cognition signals on language models. Before that, she graduated top of the cohort as a valedictorian from the University of Macau, with a B.A. in English and a minor in Computer Science.

Interests
  • Multimodal Learning
  • Prompting
  • Aligning Cognition with NLP
  • Debiasing Large Language Models
Education
  • MPhil in Linguistics (NLP track), 2022-2023

    University of Cambridge, UK

  • Funded Exchange, 2019-2020

    University of Michigan, Ann Arbor, USA

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

    University of Macau, Macao S.A.R., China

Publications

(2023). Multimodal Tuning with Human Cognition Data as Prompts. (Master’s Dissertation).

(2022). Reading-While-Listening vs. Reading-Only in A Second Language at Different Language Proficiencies: an Eye-Tracking Study. In AsiaTEFL 2022 (also presented at The LELPG Conference 2023 at the University of Edinburgh).

PDF

(2021). Pedagogy in a Pandemic: College Instructor Perspectives on Online Instruction during COVID-19 at Universities in USA and China. In AERA 2022 Annual Meeting.

Link

Research Experience

 
 
 
 
 
Language Technology Lab (LTL), Uniersity of Cambridge
Research Student
February 2022 – Present Cambridge, UK

Topic: Enriching Language Models with Human Cognition Data via Multi-Modal Prompt-based Finetuning.

Content:

  • Explored both hard and soft prompting methods to introduce cognition signals into SOTA language models;
  • Designed a multimodal mapping framework ‘CogMAP’ (Cognition Mapping And Prompting);
  • Acquired a robustly large performance gain with both eye-tracking and EEG signals on CogMAP.
 
 
 
 
 
Centre for Cognitive and Brain Sciences, Uniersity of Macau
Research Assistant
January 2021 – February 2022 Macao S.A.R, China

Topic: Examining Bilinguals’ Cognitive Attention to Content Words During Listening Assessments.

Content:

  • Used Praat to locate and extract the trial-specific speech data duration;
  • Wrote code to align temporal gaze data with speech data, extracting word-level fixation hits;
  • Analysed the correlation between gaze variables and L2 proficiency level.
 
 
 
 
 
Department of Computer Science & of Psychology, University of Macau
Research Intern
July 2019 – December 2020 Macao S.A.R, China

Topic: Applying NLP in Developmental Psychology.

Content:

  • Dataset collection and annotation;
  • Trained a Bi-LSTM on sequence classification of growth mindsets;
  • Analysed individuals’ growth mindsets and socio-economic status in correlation with their lexical usage patterns.
 
 
 
 
 
Department of Education & of Psychology, University of Michigan
Research Student (Hybrid)
April 2020 – July 2021 Ann Arbor, MI, USA
Topic: A Cross-Cultural Survey on College Instructor Experiences with Online Instruction during COVID-19 at Universities in USA and China.

Accomplish­ments

Macau Foundation
Fundação Macau Academic Prize
MOP 10,000 (equivalent of GBP 1,000)
Represented the graduating class of 2021 at the congregation speech.
My very first dip into NLP that sparked my career passion.
Full waiver of tuition & college accomodation fees during four academic years, overseas exchange covered.

Contact

  • yw611@cam.ac.uk
  • Murray Edwards College, Huntingdon Road, Cambridge, Cambridgeshire CB3 0DF