paper-with-me

홈 › Papers

TransMamba: Flexibly Switching between Transformer and Mamba

2025-03-31 · Yixing Li, Ruobing Xie, Zhen Yang, Xingwu Sun, Shuaipeng Li, Weidong Han, Zhanhui Kang, Yu Cheng, Chengzhong Xu, Di Wang, Jie Jiang

Transformers are the cornerstone of modern large language models, but their quadratic computational complexity limits efficiency in long-sequence processing. Recent advancements in Mamba, a state space model (SSM) with linear complexity, offer promising efficiency gains but suffer from unstable contextual learning and multitask generalization. This paper proposes TransMamba, a novel framework that unifies Transformer and Mamba through shared parameter matrices (e.g., QKV and CBx), and thus could dynamically switch between attention and SSM mechanisms at different token lengths and layers. We design the Memory converter to bridge Transformer and Mamba by converting attention outputs into SSM-compatible states, ensuring seamless information flow at TransPoints where the transformation happens. The TransPoint scheduling is also thoroughly explored for further improvements. We conducted extensive experiments demonstrating that TransMamba achieves superior training efficiency and performance compared to baselines, and validated the deeper consistency between Transformer and Mamba paradigms, offering a scalable solution for next-generation sequence modeling.

📄 PDF Abstract BibTeX arXiv:2503.24067

Code (0)

등록된 구현이 없습니다.

Tasks

MambaScheduling

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Multi-Head Attention 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Residual Connection 설명 없음

Similar Papers 제목 키워드 기반

TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba

2025-02-21 · Xiuwei Chen, Sihao Lin, Xiao Dong, Zisheng Chen 외

Transformers have been favored in both uni-modal and multi-modal foundation models for their flexible scalability in attention modules. Consequently, a number of pre-trained Transformer models, e.g., LLaVA, CLIP, and DEI…

image-classificationImage ClassificationMambaQuestion Answering+3

A Hybrid Transformer-Mamba Network for Single Image Deraining

2024-08-31 · Shangquan Sun, Wenqi Ren, Juxiang Zhou, Jianhou Gan 외

Existing deraining Transformers employ self-attention mechanisms with fixed-range windows or along channel dimensions, limiting the exploitation of non-local receptive fields. In response to this issue, we introduce a no…

MambaRain RemovalSingle Image Deraining

LLEMamba: Low-Light Enhancement via Relighting-Guided Mamba with Deep Unfolding Network

2024-06-03 · Xuanqi Zhang, Haijin Zeng, Jinwang Pan, Qiangqiang Shen 외

Transformer-based low-light enhancement methods have yielded promising performance by effectively capturing long-range dependencies in a global context. However, their elevated computational demand limits the scalability…

Mamba

Hunyuan-TurboS: Advancing Large Language Models through Mamba-Transformer Synergy and Adaptive Chain-of-Thought

2025-05-21 · Ao Liu, Botong Zhou, Can Xu, Chayse Zhou 외

As Large Language Models (LLMs) rapidly advance, we introduce Hunyuan-TurboS, a novel large hybrid Transformer-Mamba Mixture of Experts (MoE) model. It synergistically combines Mamba's long-sequence processing efficiency…

ChatbotInstruction FollowingMambaMixture-of-Experts

Samba: A Unified Mamba-based Framework for General Salient Object Detection

2025-01-01 · CVPR 2025 1 · Jiahao He, Keren Fu, Xiaohong Liu, Qijun Zhao

Existing salient object detection (SOD) models primarily resort to convolutional neural networks (CNNs) and Transformers. However, the limited receptive fields of CNNs and quadratic computational complexity of transf…

Mambaobject-detectionObject DetectionSalient Object Detection