Zhenghai Xue's picture

3 3

Zhenghai Xue

ZhenghaiXue

·

AI_Defender

AI & ML interests

Reinforcement Learning

Recent Activity

authored a paper about 14 hours ago

SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning

authored a paper about 20 hours ago

AgentStudio: A Toolkit for Building General Virtual Agents

upvoted a paper 3 days ago

SimpleTIR: End-to-End Reinforcement Learning for Multi-Turn Tool-Integrated Reasoning

View all activity

Organizations

Collections 1

Papers 3

arxiv:2509.02479

arxiv:2505.10978

arxiv:2403.17918

models 4

ZhenghaiXue/gigpo_qwen2.5_3b_sim0.3_step150

3B • Updated Jul 30 • 9

ZhenghaiXue/gigpo_qwen2.5_3b_sim0.5_step150

3B • Updated Jul 30 • 9

ZhenghaiXue/Qwen2.5-7B-SimpleTIR

Reinforcement Learning • 8B • Updated Jul 8 • 12

ZhenghaiXue/Qwen2.5-32B-SimpleTIR

33B • Updated Jul 8 • 9

datasets 0

None public yet