cv

Education, teaching and service, publications, and talks.

General Information

Full Name Nan Li
Position PhD Candidate, Department of Information and Computing Sciences, Utrecht University
Project DeMeVaConflicting Interpretations in Dialogue
Advisors Prof. Massimo Poesio and Prof. Albert Gatt
Email n.li@uu.nl
Languages Chinese (native), English (proficient), French, Dutch and Russian (basic)
Funding Dutch Research Council (NWO), AiNed Fellowship Grant

Education

  • 2024 – present
    PhD in Computational Linguistics and Natural Language Processing
    Utrecht University, Utrecht, The Netherlands
  • 2021 – 2024
    Master in Chinese Linguistics (Chinese Information Processing)
    Peking University, Beijing, China
  • 2017 – 2021
    Bachelor in Computational Linguistics
    Peking University, Beijing, China

Teaching and Service

  • Oct 2026
    Local Organizer
    INLG 2026, the 19th International Conference on Natural Language Generation, Utrecht, The Netherlands — 17–21 October
  • 2024 – present
    Instructor, Applications of Machine Learning
    Utrecht University, Department of Information and Computing Sciences
    • Tutorial sessions, academic years 2024–2025 and 2025–2026.
    • INFOB3APML, Bachelor 3, 7.5 EC, Period 2.
  • 2022 – 2023
    Technical Assistant
    Center for Chinese Linguistics (CCL), Peking University
    • Maintained and curated the CCL Corpus, one of the most widely used Chinese language corpora.
    • Contributed to the SpaCE corpus.

Publications

Talks and Presentations

  • Aug 2026
    Personal Interpretations as Pragmatic Label Variation: Perspectivist Grounding Annotation and (V)LLM Evaluation
    HLV-DP workshop (Human Label Variation in Discourse and Pragmatic phenomena), ESSLLI 2026, Prague, Czech Republic — long talk, 12 August
  • Aug 2026
    Seeing Is Not Sharing: Some Vision-Language Models Overestimate Common Ground in Asymmetric Dialogue
    SIGDIAL 2026, Emory University, Atlanta, GA, USA — oral, 5 August
  • May 2026
    Grounded Misunderstandings in Asymmetric Dialogue: A Perspectivist Annotation Scheme for MapTask
    LREC 2026, Palma de Mallorca, Spain — oral
  • Nov 2025
    DeMeVa at LeWiDi-2025: Modeling Perspectives with In-Context Learning and Label Distribution Learning
    4th Workshop on Perspectivist Approaches to NLP (NLPerspectives), EMNLP 2025, Suzhou, China — poster
  • Oct 2025
    Grounded Misunderstandings in Asymmetric Dialogue: A Perspectivist Annotation Scheme for MapTask
    DeMeVa Year 3 Workshop, Utrecht University

Selected Awards and Scholarships

  • 2023
    • China National Scholarship, Ministry of Education
    • Merit Student (14%), Peking University
  • 2021 – 2024
    • First Class Graduate Student Academic Scholarship, Peking University
  • 2022 / 2023
    • Specialized Graduate Student Academic Scholarship, Peking University
  • 2022
    • Excellent Student Leader (1%), Peking University
  • 2021
    • Excellent Graduate, Beijing
    • Weiming Scholar, Peking University
    • Excellent Graduate, Peking University
  • 2020
    • Merit Student Pacesetter (2%), Peking University

Academic Interests

  • Dialogue, grounding, and misunderstanding
    • How interlocutors establish common ground in asymmetric collaborative tasks, and how apparent agreement can mask referential misalignment.
    • Annotation schemes that record speaker intent and addressee interpretation separately.
  • Perspectivism and human label variation
    • Modeling disagreement between annotators as signal rather than noise.
    • Label distribution learning and in-context approaches to perspective modeling.
  • Language and vision-language models
    • Whether models track how grounding develops in a dialogue or rely on static referential cues.
    • Evaluation design for multimodal reasoning about shared understanding.