Scholar
Belinda Z. Li
Google Scholar ID: 700fyvEAAAAJ
MIT
Natural Language Processing
Machine Learning
AI
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
3,558
H-index
13
i10-index
17
Publications
20
Co-authors
21
list available
Contact
Email
bzl@mit.edu
CV
Open ↗
Twitter
Open ↗
GitHub
Open ↗
Publications
3 items
Training Language Models to Explain Their Own Computations
2025
Cited
0
QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?
2025
Cited
0
(How) Do Language Models Track State?
2025
Cited
0
Resume (English only)
Academic Achievements
Recipient of NDSEG Fellowship and Clare Boothe Luce Graduate Fellowship
Published multiple papers at top venues including ACL, ICLR, NAACL, EMNLP, HRI, and ArXiv, such as:
“Eliciting Human Preferences with Language Models” (ICLR 2025)
“Implicit Representations of Meaning in Neural Language Models” (ACL 2021)
“QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?” (ArXiv)
“(How) Do Language Models Track State” (ArXiv)
“Bayesian Preference Elicitation with Language Models” (ArXiv)
Several papers list her as equal contribution (* marked)
Co-authors
21 total
Sinong Wang
Research Scientist, MSL TBD Lab, Meta
Co-author 2
Madian Khabsa
GenAI, Meta
Jacob Andreas
MIT
Scott Wen-tau Yih
Meta FAIR
Co-author 6
Srini Iyer
FAIR
Sewon Min
UC Berkeley EECS & Allen Institute for AI
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up