# Mtextsuperscript4Fuse: Lightweight State-Space MoE with a Cross-Scale Gating Bridge for Brain Tumor Segmentation

arXiv: https://arxiv.org/abs/2605.02444 · HTML: https://mlsanity.com/paper/2605.02444/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical and biomedical MoE, one of 12 directions in Mixture-of-experts.
- Builds on 20 of its 46 references that are in the map.
- First posted on arXiv 4 May 2026.

## Placed in

- Mixture-of-experts / [Medical and biomedical MoE](https://mlsanity.com/moe/medical-and-biomedical-moe/)

## Builds on (most cited first, 20 of 20)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation](https://mlsanity.com/paper/1606.06650/) (2016)
- [3D MRI brain tumor segmentation using autoencoder regularization](https://mlsanity.com/paper/1810.11654/) (2018)
- [SegMamba: Long-range Sequential Modeling Mamba For 3D Medical Image Segmentation](https://mlsanity.com/paper/2401.13560/) (2024)
- [TransBTS: Multimodal Brain Tumor Segmentation Using Transformer](https://mlsanity.com/paper/2103.04430/) (2021)
- [Deep Multimodal Fusion by Channel Exchanging](https://mlsanity.com/paper/2011.05005/) (2020)
- [MoE-Lightning: High-Throughput MoE Inference on Memory-constrained GPUs](https://mlsanity.com/paper/2411.11217/) (2024)
- [Dynamic Multimodal Fusion](https://mlsanity.com/paper/2204.00102/) (2022)
- [LightM-UNet: Mamba Assists in Lightweight UNet for Medical Image Segmentation](https://mlsanity.com/paper/2403.05246/) (2024)
- [MSRF-Net: A Multi-Scale Residual Fusion Network for Biomedical Image Segmentation](https://mlsanity.com/paper/2105.07451/) (2021)
- [PaddleSeg: A High-Efficient Development Toolkit for Image Segmentation](https://mlsanity.com/paper/2101.06175/) (2021)
- [All-In-One Medical Image Restoration via Task-Adaptive Routing](https://mlsanity.com/paper/2405.19769/) (2024)
- [XMoE: Sparse Models with Fine-grained and Adaptive Expert Selection](https://mlsanity.com/paper/2403.18926/) (2024)
- [Lightweight Real-time Semantic Segmentation Network with Efficient Transformer and CNN](https://mlsanity.com/paper/2302.10484/) (2023)
- [CSWin-UNet: Transformer UNet with Cross-Shaped Windows for Medical Image Segmentation](https://mlsanity.com/paper/2407.18070/) (2024)
- [Narrowing the semantic gaps in U-Net with learnable skip connections: The case of medical image segmentation](https://mlsanity.com/paper/2312.15182/) (2023)
- [Maximum Score Routing For Mixture-of-Experts](https://mlsanity.com/paper/2508.12801/) (2025)
- [An Overview of Load Balancing in HetNets: Old Myths and Open Problems](https://mlsanity.com/paper/1307.7779/) (2013)
- [THOR-MoE: Hierarchical Task-Guided and Context-Responsive Routing for Neural Machine Translation](https://mlsanity.com/paper/2505.14173/) (2025)
- [DCL-SE: Dynamic Curriculum Learning for Spatiotemporal Encoding of Brain Imaging](https://mlsanity.com/paper/2511.15151/) (2025)

## Abstract (arXiv)

Encoder-decoder imbalance and the reliance on large input volumes make many 3D brain tumor segmentation models both compute-heavy and brittle. We present Mtextsuperscript4Fuse, a lightweight network that prioritizes discriminative brain tumor cues over exhaustive appearance reconstruction. Our method balances encoder and decoder capacity and replaces depth expansion with a synergistic design: it propagates long-range context with linear complexity via a grouped state space mixer, denoises and aligns skip features using a cross-scale dual-stage gating bridge, and absorbs cross-site acquisition shifts with a sample-level mixture-of-experts. On the BraTS2019 and BraTS2021 benchmarks, Mtextsuperscript4Fuse outperforms other lightweight excellent methods in both parameter count and performance. Even at a challenging input resolution of (64×128×128) (half that of existing excellent models), Mtextsuperscript4Fuse reduces parameters by 62.63% and improves average performance by 0.09%. Ablations of key components validate the method's exceptional parameter-to-accuracy efficiency and robustness across diverse data centers.