// ∿ New Architectures: SSM & Linear Attention
Mamba-2 (State Space Duality)
Shows SSMs and a class of attention are dual, yielding faster Mamba.
Mamba-2 / 状态空间对偶 2024 Research frontier extended
IN PLAIN WORDS
Introduces state space duality: selective SSMs equal masked matrix multiplications, enabling tensor-core-friendly computation and larger states.
Used in: Hybrid-architecture LLMs
TIMELINE · prerequisites → this → what builds on it
01Original paper
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
Tri Dao, Albert Gu · 2024
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Then you can learn
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
Reference code
04Where the course teaches it
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
mamba2)