Linxi Zhao
lz586 at cornell dot edu
I’m a rising third-year Ph.D. student in Computer Science at
Cornell,
focusing on machine learning, particularly large language models.
I am fortunate to be advised by
Prof. Kilian Weinberger
and Prof. Jennifer J. Sun, and I also work closely with Prof. Yoav Artzi. I'm honored to be a recipient of the LinkedIn PhD Fellowship. This summer, I'm a research intern at Apple Machine Learning Research.
Previously, I earned my B.E. from Xinya College, the liberal arts college at Tsinghua. I also spent a summer at UCLA for a research internship.
I’m currently interested in building modular, sparse, and specialized LLMs, especially exploring new forms of memory as a new scaling axis and unlocking capabilities beyond today's monolithic dense models. Along this line, our LMLM work and its follow-up Co-LMLM externalize factual knowledge into an explicit, editable knowledge base rather than memorizing it in model weights.
Outside of research, I'm an unprofessional volleyball player and an amateur photographer. I'm always happy to connect, so feel free to reach out for a chat or collaboration.
Previously, I earned my B.E. from Xinya College, the liberal arts college at Tsinghua. I also spent a summer at UCLA for a research internship.
I’m currently interested in building modular, sparse, and specialized LLMs, especially exploring new forms of memory as a new scaling axis and unlocking capabilities beyond today's monolithic dense models. Along this line, our LMLM work and its follow-up Co-LMLM externalize factual knowledge into an explicit, editable knowledge base rather than memorizing it in model weights.
Outside of research, I'm an unprofessional volleyball player and an amateur photographer. I'm always happy to connect, so feel free to reach out for a chat or collaboration.
Research
Pre-training Limited Memory Language Models with Internal and External Knowledge
Linxi Zhao, Sofian Zalouk, Christian K. Belardi, Justin Lovelace, Jin Peng Zhou, Ryan Thomas Noonan, Dongyoung Go, Kilian Q. Weinberger, Yoav Artzi, Jennifer J. Sun
ICLR 26' (Top 2%)
CCFM @ Neurips 25' Best Paper Runner-up
[Arxiv]
|
[Project Page]
|
[Github]
|
[Talk]
LMLM is pronounced as LamLam 🦙.
Co-Evolving Structured Knowledge and Reasoning in Language Models
Ryan Thomas Noonan*, Linxi Zhao*†, Menghan Xu*, Akanksha Sarkar, Mihir Mishra, Dongyoung Go, Kilian Q. Weinberger, Yoav Artzi, Jennifer J. Sun
COLM 26'
† project lead
Professional Services
Jul. 2025
One of the organizers for LM4Sci @ COLM 2025,
a Workshop on Large Language Modeling for Scientific Discovery.
We’d love for you to check it out!
2024 – Present
Reviewer for ICML, ICLR, NeurIPS, COLM, and CVPR
Gold Reviewer Award, ICML
Gold Reviewer Award, ICML
Teaching Experience
Aug. 2024 – Dec. 2024
Teaching Assistant for Cornell CS3780/5780 Introduction to Machine Learning
Invited Talks
Sep. 2025
Guest lecture in Cornell CS6784: Advanced Topics in Machine Learning
May. 2025
Invited talk at Xinya College, Tsinghua
Mar. 2025
Invited talk at the Caltech Foundation Models Workshop, organized by Prof. Yisong Yue
Awards
2025
LinkedIn PhD Fellowship
2024
Honors Graduate, Tsinghua University
2024
Dean's List, Xinya College
Extracurricular Activities
2024
Member of Tsinghua University Volleyball Team (Non-Athlete Track)
Gold Medalist, Beijing Intercollegiate Tournament
Gold Medalist, Beijing Intercollegiate Tournament
2021
Volunteered as a teacher in a rural village in Ningxia, China, supporting left-behind children
Tsinghua's Golden Social Practice Award
Tsinghua's Golden Social Practice Award