Summary of Famba-v: Fast Vision Mamba with Cross-layer Token Fusion, by Hui Shen et al.
Famba-V: Fast Vision Mamba with Cross-Layer Token Fusion
by Hui Shen, Zhongwei Wan, Xin Wang, Mi Zhang
First submitted to arxiv on: 15 Sep 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Famba-V, a novel token fusion technique, enhances the training efficiency of Vision Mamba (Vim) models by reducing both training time and peak memory usage during training. Unlike existing works, Famba-V identifies and fuses similar tokens across different Vim layers using cross-layer strategies, leading to superior accuracy-efficiency trade-offs on CIFAR-100. This work demonstrates Famba-V as a promising efficiency enhancement technique for Vim models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper introduces a new way to improve the training process of Vision Mamba (Vim) models, which are used in image recognition tasks. The new method, called Famba-V, makes the training faster and uses less memory by combining similar information from different layers of the model. This results in better performance for the same amount of computation. The researchers tested this method on a dataset and found it works well. |
Keywords
» Artificial intelligence » Token
