AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Mamba-ND: Selective State Space Modeling for Multi-Dimensional Data

Shufan Li, Aditya Grover, Harkanwar Singh

ECCV 2024 · Architectures and efficiency › Attention and transformers · approach: Mamba architectures · 138 citations (OpenAlex)

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 changedWhereMeasuredEffect
alternate scan orderings across layers (H+H-W+W-T+T-) instead of flattening into one 1D sequence (ablation)Mamba-ND · HMDB-51top-1 accuracy: 34.9 -> 59.0Improved
alternate scan orderings across layers instead of 1D flattening (ablation)Mamba-ND · ImageNet-1K valtop-1 accuracy: 76.4 -> 79.4Improved
alternating-directional design versus layer-level Multi-Head SSM / ND-SSM / Bi-SSM (ablation)Mamba-ND · ImageNet-1K and HMDB-51top-1 accuracy: 79.4 vs 77.6, 77.2, 74.6 on ImageNetImproved
layer-level Bi-SSM instead of single-direction 1D-SSM (ablation)Mamba-ND · ImageNet-1K valtop-1 accuracy: 76.4 -> 74.6Hurt
group directions into parallel multi-directional blocks at block level (ablation)Mamba-ND · HMDB-51top-1 accuracy: 59.0 -> 38.3 (hex-directional block)Hurt
factorize the 3D scan into smaller sequences (ablation)Mamba-ND · HMDB-51top-1 accuracy: 59.0 -> 55.8 (2D+1D)Hurt
smaller patch size 8 instead of 16 (ablation)Mamba-2D-S · ImageNet-1K valtop-1 accuracy: 79.4 -> 81.7Improved
set temporal SSM initial delta scale to 1.0 versus 5.0 (ablation)Mamba-3D · HMDB-51top-1 accuracy: 55.3 -> 59.0Improved
change ordering of first three layers or swap TWH+ to WTH+ (ablation)Mamba-3D · BTCVDICENo clear effect
Mamba-ND instead of Transformer baselinesMamba-3D · ERA5 weather forecastingACC: 89.3 -> 90.1Improved
Mamba-3D versus Swin-UNETR on BTCV with a small modelMamba-3D-UNETR-S · BTCV 3D segmentationDICE: 83.3 -> 83.1No clear effect
add alternating directions to Bi-SSM (ablation)Bi-SSM · ImageNet-1K and HMDB-51top-1 accuracy: +1.5 ImageNet, +8.4 HMDB-51Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy