Publications
* denotes equal contribution. Authorship in robust statistics works is in alphabetical order.
Language Models
2025
Can Large Language Models Develop Strategic Reasoning? Post-training Insights from Learning Chess
ScalR workshop at COLM 2025.
Alignment as Distribution Learning: Your Preference Model is Explicitly a Language Model
FoPT workshop at COLT 2025.
Lexico: Extreme KV Cache Compression via Sparse Coding over Universal Dictionaries
ICML 2025. ICLR Workshop on Sparsity in LLMs (spotlight).
Task Diversity Shortens the ICL Plateau
TMLR 2025.
Robust Statistics
GLM Regression with Oblivious Corruptions
COLT 2023.
Teaching
I have taught the following courses as a TA (*Head TA) in my undergraduate and graduate years.
UC Berkeley
UW Madison
Fa19: CS240 (Discrete Math)
Sp20, Sp21*: CS577 (Algorithms)
Fa20: CS787 (Advanced Algorithms)