Hi! I'm Xinyi.

avator.jpg
Gates 319, 3800 E Stevens Way NE, Seattle, WA 98195

[email protected]

I am a Postdoctoral Scholar in the Paul G. Allen School of Computer Science & Engineering and a Data Science Postdoctoral Fellow in the eScience Institute at the University of Washington. I am fortunate to work with Amy Zhang and Tim Althoff.

My research aims to advance AI, with a current focus on LLMs and RAG, to foster social and technological trust. It leverages multimodal and behavioral data while integrating insights from social sciences to address social challenges—ranging from m/disinformation to well-being—and to build aligned and trustworthy technologies for diverse populations. Check out our recent work as an example; feedback is welcome!

I collaborate with exceptional researchers, practitioners, and students from various disciplines and organizations (e.g., nonprofits). Our work has been published in journals such as ACM Computing Surveys (IF: 23.8) and presented at conferences, including WWW, KDD, WSDM, CIKM, ICWSM, and PAKDD, with 4200+ citations.

I was selected as a Rising Star in EECS and a Rising Star in Data Science in 2024.

Travels & Talks

Nov 14, 2024 Will attend the Rising Stars in Data Science workshop @ UCSD
Nov 05, 2024 Will present @ UW Data Science Seminar
Oct 24, 2024 Will attend the Rising Stars in EECS workshop @ MIT
Oct 11, 2024 Will visit DMML Lab @ ASU
Sep 26, 2024 Will attend the Trust & Safety Research Conference @ Stanford University and present our work—I’ll be around until the 28th; let’s chat if you’re also around and interested in LLMs, RAG, misinformation, etc.
▸ More
Feb 06, 2024 Presented our work @ UW Center for an Informed Public
Feb 01, 2024 Presented our work @ UW eScience Institute

Publications

See the full list in Google Scholar.
  1. MUSE.svg
    Correcting misinformation on social media with a large language model
    Xinyi ZhouAshish SharmaAmy X Zhang, and Tim Althoff
    arXiv preprint arXiv:2403.11169, 2024
  2. WWW
    “This is fake! Shared it by mistake”: Assessing the intent of fake news spreaders
    Xinyi ZhouKai Shu, Vir V Phoha, Huan Liu, and Reza Zafarani
    In Proceedings of the ACM Web Conference, 2022
  3. CSUR
    A survey of fake news: Fundamental theories, detection methods, and opportunities
    Xinyi Zhou, and Reza Zafarani
    ACM Computing Surveys, 2020
  4. PAKDD
    SAFE: Similarity-aware multimodal fake news detection
    Xinyi Zhou, Jindi Wu, and Reza Zafarani
    In Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2020
  5. CIKM
    RECOVERY: A multimodal repository for COVID-19 news credibility research
    Xinyi Zhou, Apurva Mulay, Emilio Ferrara, and Reza Zafarani
    In Proceedings of the 29th ACM International Conference on Information & Knowledge Management, 2020
  6. ICWSM
    Sentiment paradoxes in social networks: Why your friends are more positive than you?
    Xinyi Zhou, Shengmin Jin, and Reza Zafarani
    In Proceedings of the International AAAI Conference on Web and Social Media, 2020
  7. WSDM Tutorial
    Fake news: Fundamental theories, detection strategies and challenges
    Xinyi ZhouReza ZafaraniKai Shu, and Huan Liu
    In Proceedings of the 12th ACM International Conference on Web Search and Data Mining, 2019

Grants, Awards, & Honors

2024
  • Rising Stars in Data Science
  • Rising Stars in EECS
  • Postdoc Research Award ($9,200)
  • Before 2024
  • Center for an Informed Public Innovation Fund ($21,228, Co-PI)
  • All-University Doctoral Prize
  • Research Excellence Doctoral Funding ($10,438)
  • Varshney Scholarship
  • Top 100 Influential Research Papers of 2017 in China
  • National Scholarship
  • Services

    Program Committee Member
    2025
  • KDD, AAAI, WSDM
  • 2024
  • KDD, AAAI, WSDM, CIKM
  • Before 2024
  • KDD, AAAI, WSDM, CIKM, SIGIR, ECML-PKDD, ECIR, IJCNN
  • Journal Reviewer
    2024
    TOIS, TWEB
    Before 2024
    TOIS, TWEB, CSUR, TKDE, TIFS, TMM, TNNLS, Omega

    Meet Pumpkin, my cat 🤗

    I took so many photos of Pumpkin and recently started moving them to Instagram (@pumpkin.not.pumpkout).