# Core SSM architectures for language

Area: [State-space sequence models (S4, Mamba)](https://mlsanity.com/ssm/index.md). HTML: https://mlsanity.com/ssm/core-ssm-architectures-for-language/

## Key facts

- Core SSM architectures for language: 278 papers on arXiv since 2021 (data through 14 Oct 2026).
- 152 of its 278 mapped papers (55%) cite “Mamba: Linear-Time Sequence Modeling with Selective State Spaces” (2023, arXiv 2312.00752).
- 114 of its 278 mapped papers (41%) cite “Efficiently Modeling Long Sequences with Structured State Spaces” (2021, arXiv 2111.00396).
- 14 new papers 18 Sep–14 Oct.

## Most cited here

- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023), cited by 152 papers here
- [Efficiently Modeling Long Sequences with Structured State Spaces](https://mlsanity.com/paper/2111.00396/) (2021), cited by 114 papers here
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024), cited by 99 papers here
- [Simplified State Space Layers for Sequence Modeling](https://mlsanity.com/paper/2208.04933/) (2022), cited by 53 papers here
- [Jamba: A Hybrid Transformer-Mamba Language Model](https://mlsanity.com/paper/2403.19887/) (2024), cited by 40 papers here
- [Diagonal State Spaces are as Effective as Structured State Spaces](https://mlsanity.com/paper/2203.14343/) (2022), cited by 33 papers here
- [An Empirical Study of Mamba-based Language Models](https://mlsanity.com/paper/2406.07887/) (2024), cited by 30 papers here
- [The Hidden Attention of Mamba Models](https://mlsanity.com/paper/2403.01590/) (2024), cited by 25 papers here
- [Gated Linear Attention Transformers with Hardware-Efficient Training](https://mlsanity.com/paper/2312.06635/) (2023), cited by 23 papers here
- [Can Mamba Learn How to Learn? A Comparative Study on In-Context Learning Tasks](https://mlsanity.com/paper/2402.04248/) (2024), cited by 22 papers here

## New (18 Sep–14 Oct)

- [TurboMamba: Faster Inference, Faster Training, and Better Prediction with Lexical Memory](https://mlsanity.com/paper/2610.20682/) (2026-10-14)
- [Beyond State Capacity: Why Retrieval Breaks in Mamba](https://mlsanity.com/paper/2610.19709/) (2026-10-14)
- [SuPER-Mamba: Surprisal Prioritization for Extended-Context Routing in Mamba](https://mlsanity.com/paper/2610.16636/) (2026-10-12)
- [A Sovereign, Indic-First, Reasoning Model](https://mlsanity.com/paper/2610.15371/) (2026-10-11)
- [READ: Reading the Selective Scan for Faithful Video Attribution](https://mlsanity.com/paper/2610.13776/) (2026-10-09)
- [Understanding Mamba-3 Through Oscillatory Dynamics](https://mlsanity.com/paper/2610.13341/) (2026-10-09)
- [Distributed Learning with Selective State Space Models: Architecture-Aware Convergence Analysis](https://mlsanity.com/paper/2610.02659/) (2026-10-02)
- [Low-Discrepancy Dither for Quantized Recurrent State Caches](https://mlsanity.com/paper/2609.39185/) (2026-09-30)
- [SketchSSM: Write to the Full State, Read from a Compact Sketch](https://mlsanity.com/paper/2609.33051/) (2026-09-27)
- [Elastic Selective Spectral Hybrids for Train-Once, Export-Many Budgeted Inference](https://mlsanity.com/paper/2609.32486/) (2026-09-26)
- [Residual Streams Read, Recurrent States Remember: The Global Workspace in Mamba Models](https://mlsanity.com/paper/2609.32102/) (2026-09-26)
- [Non-Commutative State Tracking with Input-Dependent Low-Rank Updates in Mamba-3](https://mlsanity.com/paper/2609.28273/) (2026-09-23)
- [Comparing Latent Concept Formation in State Space Models and Transformers via Sparse Autoencoders](https://mlsanity.com/paper/2609.24440/) (2026-09-21)
- [H-Spec: Parallel Speculative Decoding Without a Drafter-Side KV Cache](https://mlsanity.com/paper/2609.24197/) (2026-09-21)