paper-with-me

홈 › Papers

DeciMamba: Exploring the Length Extrapolation Potential of Mamba

2024-06-20 · Assaf Ben-Kish, Itamar Zimerman, Shady Abu-Hussein, Nadav Cohen, Amir Globerson, Lior Wolf, Raja Giryes

Long-range sequence processing poses a significant challenge for Transformers due to their quadratic complexity in input length. A promising alternative is Mamba, which demonstrates high performance and achieves Transformer-level capabilities while requiring substantially fewer computational resources. In this paper we explore the length-generalization capabilities of Mamba, which we find to be relatively limited. Through a series of visualizations and analyses we identify that the limitations arise from a restricted effective receptive field, dictated by the sequence length used during training. To address this constraint, we introduce DeciMamba, a context-extension method specifically designed for Mamba. This mechanism, built on top of a hidden filtering mechanism embedded within the S6 layer, enables the trained model to extrapolate well even without additional training. Empirical experiments over real-world long-range NLP tasks show that DeciMamba can extrapolate to context lengths that are significantly longer than the ones seen during training, while enjoying faster inference.

📄 PDF Abstract BibTeX arXiv:2406.14528

Code (2)

assafbk/decimamba 공식 구현 pytorch
nadavsc/diff-mamba pytorch

Tasks

Mamba

Similar Papers 제목 키워드 기반

Recall with Reasoning: Chain-of-Thought Distillation for Mamba's Long-Context Memory and Extrapolation

2025-05-06 · Junyu Ma, Tianqing Fang, Zhisong Zhang, Hongming Zhang 외

Mamba's theoretical infinite-context potential is limited in practice when sequences far exceed training lengths. This work explores unlocking Mamba's long-context memory ability by a simple-yet-effective method, Recall …

Mamba

DocMamba: Efficient Document Pre-training with State Space Model

2024-09-18 · Pengfei Hu, Zhenrong Zhang, Jiefeng Ma, Shuhang Liu 외

In recent years, visually-rich document understanding has attracted increasing attention. Transformer-based pre-trained models have become the mainstream approach, yielding significant performance gains in this field. Ho…

document understanding

State-space models are accurate and efficient neural operators for dynamical systems

2024-09-05 · Zheyuan Hu, Nazanin Ahmadi Daryakenari, Qianli Shen, Kenji Kawaguchi 외

Physics-informed machine learning (PIML) has emerged as a promising alternative to classical methods for predicting dynamical systems, offering faster and more generalizable solutions. However, existing models, including…

Computational EfficiencyMambaOperator learningPhysics-informed machine learning+1

StruMamba3D: Exploring Structural Mamba for Self-supervised Point Cloud Representation Learning

2025-06-26 · Chuxin Wang, Yixin Zha, Wenfei Yang, Tianzhu Zhang

Recently, Mamba-based methods have demonstrated impressive performance in point cloud representation learning by leveraging State Space Model (SSM) with the efficient context modeling ability and linear complexity. Howev…

MambaRepresentation Learning

VL-Mamba: Exploring State Space Models for Multimodal Learning

2024-03-20 · Yanyuan Qiao, Zheng Yu, Longteng Guo, Sihan Chen 외

Multimodal large language models (MLLMs) have attracted widespread interest and have rich applications. However, the inherent attention mechanism in its Transformer structure requires quadratic complexity and results in …

Language ModelingLanguage ModellingLarge Language ModelMamba+3