Xiangyu's picture

16 16

Xiangyu

xixy

·

https://xixy.github.io/

AI & ML interests

None yet

Recent Activity

upvoted a paper 12 days ago

Universal Reasoning Model

commented on a paper 13 days ago

Rethinking Expert Trajectory Utilization in LLM Post-training

commented on a paper 13 days ago

State over Tokens: Characterizing the Role of Reasoning Tokens

View all activity

Organizations

None yet

authored a paper 7 months ago

Rethinking the Sampling Criteria in Reinforcement Learning for LLM Reasoning: A Competence-Difficulty Alignment Perspective

Paper • 2505.17652 • Published May 23 • 6

authored a paper 10 months ago

SampleMix: A Sample-wise Pre-training Data Mixing Strategey by Coordinating Data Quality and Diversity

Paper • 2503.01506 • Published Mar 3 • 10