用可学习注意力机制提升医学图像分割的跨阶段信息融合
XAttnRes: Cross-Stage Attention Residuals for Medical Image Segmentation
- 引入跨阶段注意力残差，动态聚合编码器与解码器所有历史特征
- 在四个数据集上均显著提升分割性能，且无需传统跳跃连接仍达基线水平
- 轻量设计适配多尺度结构，适合医疗图像分割模型改进
在大语言模型中，注意力残差已被证明能通过学习方式选择性聚合所有前序层输出，优于固定残差连接。本文提出跨阶段注意力残差（XAttnRes），构建一个全局特征历史池，累积编码器与解码器各阶段输出。通过轻量级伪查询注意力，每个阶段可选择性地从全部前置表示中聚合信息。为弥合大语言模型中同维变换层与分割网络多尺度编解码阶段之间的差距，XAttnRes引入空间对齐与通道投影步骤，以极低开销处理跨分辨率特征。将XAttnRes加入现有分割网络后，在四个数据集和三种成像模态上均持续提升性能。进一步观察发现，仅使用XAttnRes而无跳跃连接时，性能已与基线相当，表明学习到的聚合机制可恢复传统预设连接所提供的跨阶段信息流。
原文摘要 · Abstract (English)
In the field of Large Language Models (LLMs), Attention Residuals have recently demonstrated that learned, selective aggregation over all preceding layer outputs can outperform fixed residual connections. We propose Cross-Stage Attention Residuals (XAttnRes), a mechanism that maintains a global feature history pool accumulating both encoder and decoder stage outputs. Through lightweight pseudo-query attention, each stage selectively aggregates from all preceding representations. To bridge the gap between the same-dimensional Transformer layers in LLMs and the multi-scale encoder-decoder stages in segmentation networks, XAttnRes introduces spatial alignment and channel projection steps that handle cross-resolution features with negligible overhead. When added to existing segmentation networks, XAttnRes consistently improves performance across four datasets and three imaging modalities. We further observe that XAttnRes alone, even without skip connections, achieves performance on par with the baseline, suggesting that learned aggregation can recover the inter-stage information flow traditionally provided by predetermined connections.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。