This repository serves as a curated, community-driven collection of academic papers focused on Mamba and State Space Models (SSMs). It aims to provide researchers and practitioners with a comprehensive and up-to-date resource for exploring advancements in this rapidly evolving area of sequence modeling, offering a valuable starting point for understanding and contributing to the field.

How It Works

The repository organizes papers by application domain (e.g., Vision, Language, Medical, Time Series) and includes links to their respective arXiv pages and, where available, associated code repositories. It is actively maintained through community contributions, encouraging users to submit new papers via issues or email, ensuring the list remains current with the latest research.

Quick Start & Requirements

This repository is a curated list of papers and does not require installation or execution. It serves as a reference guide.

Highlighted Details

• Comprehensive categorization of Mamba and SSM research across diverse domains including Vision, Language, Medical Imaging, Time Series, Speech, Graphs, Point Clouds, Multimodal learning, Reinforcement Learning, Diffusion Models, Embodied AI, and Healthcare.
• Includes links to over 200 papers, many with associated code repositories, facilitating practical exploration.
• Regularly updated with recent publications, including those accepted at major conferences like CVPR, ECCV, ICML, and MICCAI.
• Provides links to supplementary resources such as surveys, explanations, and other curated lists for deeper learning.

Maintenance & Community

The repository is actively maintained by the community, with a clear invitation for contributions via GitHub issues or direct email. The primary contributor is Yujin Tang.

Licensing & Compatibility

The repository itself is not software and therefore not subject to software licensing. The content consists of links to academic papers, which are typically governed by their own publication licenses (e.g., arXiv's terms).

Limitations & Caveats

This repository is a curated list and does not provide implementations or benchmarks. The quality and relevance of the papers are subject to community curation.