xuxin's picture

On Vacation 🏝️

3 9

xuxin

xx18

·

https://xinxu-ustc.github.io/

AI & ML interests

None yet

Recent Activity

upvoted a paper 11 days ago

ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestration

upvoted a paper 23 days ago

EntroPIC: Towards Stable Long-Term Training of LLMs via Entropy Stabilization with Proportional-Integral Control

updated a model about 1 month ago

xx18/TFPI-Qwen3-4B-Thinking-2507-Stage3

View all activity

Organizations

commented 2 papers 2 months ago

On Predictability of Reinforcement Learning Dynamics for Large Language Models

Paper • 2510.00553 • Published Oct 1 • 8 •

Thinking-Free Policy Initialization Makes Distilled Reasoning Models More Effective and Efficient Reasoners

Paper • 2509.26226 • Published Sep 30 • 32 •

New activity in UGMathBench/ugmathbench 10 months ago

Update link to paper, add Github repo

#2 opened 10 months ago by