Microsoft's Universal YOCO enables efficient depth scaling
MSRA researchers published Universal YOCO, using shared-parameter recursion for efficient depth scaling. The YOCO architecture also evolves into a family with a new sparse-attention branch alongside Looped YOCO.
2026-09-11 ~ 2026-09-11 · 3 related posts
- MSRA team unveils Universal YOCO: shared-parameters recursion for efficient depth scaling — donglixp · 2026-09-11
- YOCO architecture expands with a Sparse Attention variant, becoming an architecture family — donglixp · 2026-09-11
- Universal YOCO paper combines recursive compute with efficient attention for depth scaling — donglixp · 2026-09-11