Andrew Zhao's picture

Andrew Zhao

andrewzh

·

https://andrewzh112.github.io/

AI & ML interests

Reinforcement Learning, Agents

Recent Activity

upvoted a paper 5 days ago

Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning

authored a paper 10 days ago

Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning

authored a paper 10 days ago

ExCyTIn-Bench: Evaluating LLM agents on Cyber Threat Investigation

View all activity

Organizations

commented a paper 9 months ago

Absolute Zero: Reinforced Self-play Reasoning with Zero Data

Paper • 2505.03335 • Published May 6, 2025 • 189 •

New activity in berkeley-nest/Starling-LM-7B-alpha about 2 years ago

Code/paper release

#26 opened about 2 years ago by