Millicent Li
Scholar

Millicent Li

Google Scholar ID: UZ1gBvAAAAAJ
Northeastern University
natural language processinghuman-computer interactioninterpretability
Citations & Impact
All-time
Citations
675
 
H-index
5
 
i10-index
4
 
Publications
7
 
Co-authors
0
 
Resume (English only)
Academic Achievements
  • Publications: 'Do Natural Language Descriptions of Model Activations Convey Privileged Information?' (arXiv, 2025); 'Multi-Field Adaptive Retrieval' accepted to ICLR 2025 as a spotlight (top 5%); 'Function Vectors in Large Language Models' accepted to ICLR 2024; 'Summarizing, Simplifying, and Synthesizing Medical Evidence using GPT-3 (with Varying Success)' accepted to ACL 2023; Awards: 2022 NSF Graduate Research Fellowship; Honorable Mention for the 2021 NSF Graduate Research Fellowship competition.
Research Experience
  • AI Resident at FAIR/Meta AI, working with Marjan Ghazvininejad and Mike Lewis; Intern at Microsoft Research, collaborating with Tristan Naumann; Undergraduate research at the University of Washington with Shwetak Patel (ubiquitous computing) and Noah Smith (natural language processing).
Education
  • PhD: Northeastern University, Advisor: Byron Wallace, Started: Fall 2022; Bachelor's Degree: University of Washington, Advisors: Shwetak Patel and Noah Smith.
Background
  • Research Interests: Language model behaviors; Specialization: Natural Language Processing, Machine Learning; Brief Introduction: A PhD student at Northeastern University, advised by Byron Wallace. Focuses on how language models can generalize precisely on unseen data and why they pick up certain behaviors.
Miscellany
  • Other personal interests or hobbies are not explicitly mentioned.
Co-authors
0 total
Co-authors: 0 (list not available)