arXiv preprint ·
FLASHSWIN: Unlocking Large Windows and Dense Tokens in Swin Vision Transformers with Memory Efficient Attention
- Published
- Source
- arXiv preprint, posted before peer review
- Original
- Read it at arXiv
- Topics
Abstract
High-resolution vision backbones have long been forced to trade away local token density to afford larger receptive fields. Hierarchical Swin transformers impose this compromise because standard windowed attention materializes an $M^2\times M^2$ score matrix per window, incurring $O(M^4)$ memory as windows or token grids grow. Furthermore, Swin adds a learned relative-position bias elementwise to attention scores, requiring full materialization of the score matrix and its gradient. This keeps Swin and SwinV2 trapped in a small-window($M=8,16$), coarse-token regime with patch size $4\times4$ ($p=4$), limiting performance for fine-grained tasks. We introduce FLASHSWIN, which replaces standard windowed attention with a FlashAttention implementation that computes exact softmax attention without materializing the score matrix, reducing per-window memory from $O(M^4)$ to $O(M^2)$. This enables higher token density and larger receptive fields without inflating memory overhead. Training memory is flat across window sizes: at a $32\times32$ window, FLASHSWIN-T requires only $12.4$\,GB, unchanged from $8\times8$, compared to $70/90$\,GB for SwinV2/V1-T. However, applying FlashAttention directly to Swin creates a trade-off: bypassing the score matrix precludes Swin's additive relative-position bias, forfeiting spatial information in exchange for memory efficiency. FLASHSWIN restores position information as window-local learnable 2D RoPE, making large windows and dense token grids both affordable and accurate. At matched scale, FLASHSWIN-T outperforms Swin variants. With dense tokens and wide windows ($p=2,M=32$), the same Tiny model reaches $84.1\%$ ImageNet-1K, $44.1$ COCO box AP, and $47.28$ ADE20K mIoU---gains of $+1.3$, $+5.1$, and $+1.82$ over SwinV2-T at $M=16$, respectively. At fixed $M=32$, halving the patch size yields roughly $3\times$ larger gains in boundary quality than in mIoU.
The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.
Read it on ReadSearch
Each opens after you create an account, and brings you back to this paper.
More in Computer Vision
Diagnosis-Conditioned Spatial Gating and Decoder-Level Supervised Contrastive Learning for Radiology Report Generation
RSure-Agent: Reliable Use of Tool Observations for Remote Sensing Agents
Referring Multi-Object Tracking in Moving-Camera Videos via Global Motion Compensation
One Tile, Multiple Instances: Rethinking MIL for Sparse Diagnostic Evidence
CellSplat4D: PSF-Aware 4D Gaussian Splatting for Sparse Robotic Live-Cell Imaging
Get the few papers worth your time in Computer Vision every morning.
ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.