[Feature] Add off-policy sequence masking algorithm proposed in DeepSeek v3.2 (#999)
C
Chengxing Xie committed
caa4af28354e3dfc0d11156d4e3529b7eb865d9c
Parent: 31af71e
Committed by GitHub <noreply@github.com>
on 12/2/2025, 11:52:02 AM