Popular GitHub Repositories for State Space Model

[IEEE TGRS 2024] ChangeMamba: Remote Sensing Change Detection Based on Spatio-Temporal State Space Model

[NeurIPS 2024] PointMamba: A Simple State Space Model for Point Cloud Analysis

A PyTorch implementation of the paper "ZigMa: A DiT-Style Mamba-based Diffusion Model" (ECCV 2024)

[NeurIPS 2024] Official Repository of The Mamba in the Llama: Distilling and Accelerating Hybrid Models

Notes on the Mamba and the S4 model (Mamba: Linear-Time Sequence Modeling with Selective State Spaces)

Official Implementation of the work "Audio Mamba: Bidirectional State Space Model for Audio Representation Learning"

Foundation models based medical image analysis

[SLT'24] The official implementation of SSAMBA: Self-Supervised Audio Representation Learning with Mamba State Space Model

Mamba in Vision: A Comprehensive Survey of Techniques and Applications

Imputation-based Time-Series Anomaly Detection with Conditional Weight-Incremental Diffusion Models, KDD 2023

Code and pretrained models for the paper: "MatMamba: A Matryoshka State Space Model"

LLM inference in Fortran

Neural State-Space Models and Latent Dynamics Functions in PyTorch for High-Dimensional Forecasting

Event stream based visual object tracking using Mamba/State Space Model

[PRCV-2024] State Space Model based Frame-Event Tracking

Official repository of ”Mamba-FSCIL: Dynamic Adaptation with Selective State Space Model for Few-Shot Class-Incremental Learning"

PyTorch implementation of the NCDSSM models presented in the ICML '23 paper "Neural Continuous-Discrete State Space Models for Irregularly-Sampled Time Series".

Mamba4Cast, a zero-shot time series forecasting model, achieves competitive performance and faster inference than transformer-based models by leveraging Mamba architecture and synthetic training data to generalize robustly without fine-tuning.

[NeurIPS 2024] MambaSCI: Efficient Mamba-UNet for Quad-Bayer Patterned Video Snapshot Compressive Imaging

Expanding linear RNN state-transition matrix eigenvalues to include negatives improves state-tracking tasks and language modeling without added training or inference costs.