arxiv:2601.22108
🔄 In a Training Loop
Shuqi Ke
shuqike
·
AI & ML interests
I work to close the generalization and sample efficiency gap between AI and human.
Recent Activity
liked a model 12 days ago
nvidia/GR00T-N1-2B upvoted a paper 8 months ago
Value-Based Pre-Training with Downstream Feedback submitted a paper 8 months ago
Value-Based Pre-Training with Downstream FeedbackOrganizations
datasets 5
shuqike/omega-explorative
Viewer • Updated • 52.2k • 84 • 1
shuqike/NuminaMath-CoT
Viewer • Updated • 860k • 36
shuqike/NuminaMath-CoT-decontam-noGSM8K-MATH
Viewer • Updated • 824k • 34
shuqike/ADE20K
Viewer • Updated • 27.6k • 23
shuqike/tiny-imagenet-200-clean
Viewer • Updated • 108k • 10