Archive

5 posts published so far

Oct 18, 2026
A Risk Penalty Cannot Fix a Confidently Wrong World Model
ym p
Oct 16, 2026
A World Model Can Lower MSE and Still Choose the Wrong Action
ym p
Oct 15, 2026
Constitutional AI in Python: Keep the Soft Labels, Not Just the Winner
ym p
Oct 13, 2026
TensorDrill update: visual AI coding practice and interview labs
ym p
Oct 11, 2026
Beyond LeetCode: Where to Practice AI Coding, from Attention to Flow Matching
ym p