Research map / Architectures and efficiency / Attention and transformers
MIMOMamba: From Scalar Duality to Matrix-Valued Attention
Yanbo Li, Richard Cornelius Suwandi, Feng Yin, SUN Yiyong, Wei Huang, Wenqiang Pu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace Transformer with MIMOMamba | SSP forecaster · SSP prediction | RMSE mean: 0.748 -> 0.688 | Improved |
| replace Transformer with MIMOMamba | SSP forecaster · SSP prediction | Params: 100,801 -> 35,593 | Improved |
| replace LSTM with MIMOMamba | SSP forecaster · SSP prediction | RMSE max: 2.993 -> 1.662 | Improved |
| replace Transformer with MIMOMamba on thermocline region | SSP forecaster · SSP thermocline (~30 m) | RMSE max: 14% reduction | Improved |
| compare MIMOMamba against Mamba-2 and Mamba-3 | MIMOMamba · SSP prediction (extended) | RMSE mean: 0.717 (Mamba-2), 0.715 (Mamba-3) -> 0.687 | Improved |
| use MIMOMamba on SCP1 versus SISO baselines | MIMOMamba · UEA SCP1 | accuracy: 85.5±2.9% | Improved |
| vary polynomial degree p from 2 to 8 (ablation) | MIMOMamba · SSP prediction | RMSE: p = 4 sweet spot | Mixed |
| use matrix polynomial sharing instead of independent dense A, B, C | MIMOMamba · core dynamics | parameter count: 3D^2 -> 1D^2 | Improved |
| use FFT polynomial multiplication at degree 4 | MIMOMamba · current configuration | speed | No clear effect |
| use MIMOMamba-Fast instead of Mamba-2 in pure PyTorch | MIMOMamba-Fast · Training throughput | throughput: 1.5-1.6x faster | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Mamba-3 — Mamba-3: Improved Sequence Modeling using State Space Principles (ICLR 2026) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality (ICML 2024 · similarity 0.89)
- Mamba Modulation: On the Length Generalization of Mamba Models (NeurIPS 2025 · similarity 0.89)
- Achilles' Heel of Mamba: Essential difficulties of the Mamba architecture demonstrated by synthetic data (NeurIPS 2025 · similarity 0.87)
- Theoretical Foundations of Deep Selective State-Space Models (NeurIPS 2024 · similarity 0.86)
- Demystifying the Token Dynamics of Deep Selective State Space Models (ICLR 2025 · similarity 0.86)
- Block-Biased Mamba for Long-Range Sequence Processing (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
