# LoMix: Learnable Weighted Multi-Scale Logits Mixing for Medical Image Segmentation

arXiv: https://arxiv.org/abs/2510.22995 · HTML: https://mlsanity.com/paper/2510.22995/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 39 references that are in the map.
- First posted on arXiv 27 Oct 2025.

## Builds on (most cited first, 27 of 27)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [UNet++: A Nested U-Net Architecture for Medical Image Segmentation](https://mlsanity.com/paper/1807.10165/) (2018)
- [Attention U-Net: Learning Where to Look for the Pancreas](https://mlsanity.com/paper/1804.03999/) (2018)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentation](https://mlsanity.com/paper/2105.05537/) (2021)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://mlsanity.com/paper/2106.13797/) (2021)
- [Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)](https://mlsanity.com/paper/1902.03368/) (2019)
- [Kvasir-SEG: A Segmented Polyp Dataset](https://mlsanity.com/paper/1911.07069/) (2019)
- [PraNet: Parallel Reverse Attention Network for Polyp Segmentation](https://mlsanity.com/paper/2006.11392/) (2020)
- [NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection](https://mlsanity.com/paper/1904.07392/) (2019)
- [UNeXt: MLP-based Rapid Medical Image Segmentation Network](https://mlsanity.com/paper/2203.04967/) (2022)
- [MedNeXt: Transformer-driven Scaling of ConvNets for Medical Image Segmentation](https://mlsanity.com/paper/2303.09975/) (2023)
- [Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformers](https://mlsanity.com/paper/2108.06932/) (2021)
- [Mixed Transformer U-Net For Medical Image Segmentation](https://mlsanity.com/paper/2111.04734/) (2021)
- [EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation](https://mlsanity.com/paper/2405.06880/) (2024)
- [UACANet: Uncertainty Augmented Context Attention for Polyp Segmentation](https://mlsanity.com/paper/2107.02368/) (2021)
- [Stepwise Feature Fusion: Local Guides Global](https://mlsanity.com/paper/2203.03635/) (2022)
- [EGE-UNet: an Efficient Group Enhanced UNet for skin lesion segmentation](https://mlsanity.com/paper/2307.08473/) (2023)
- [Fast Neural Architecture Search of Compact Semantic Segmentation Models via Auxiliary Cells](https://mlsanity.com/paper/1810.10804/) (2018)
- [Multi-scale Hierarchical Vision Transformer with Cascaded Attention Decoding for Medical Image Segmentation](https://mlsanity.com/paper/2303.16892/) (2023)
- [CMUNeXt: An Efficient Medical Image Segmentation Network based on Large Kernel and Skip Fusion](https://mlsanity.com/paper/2308.01239/) (2023)
- [G-CASCADE: Efficient Cascaded Graph Convolutional Decoding for 2D Medical Image Segmentation](https://mlsanity.com/paper/2310.16175/) (2023)
- [MK-UNet: Multi-kernel Lightweight CNN for Medical Image Segmentation](https://mlsanity.com/paper/2509.18493/) (2025)
- [An Efficient Dual-Line Decoder Network with Multi-Scale Convolutional Attention for Multi-organ Segmentation](https://mlsanity.com/paper/2508.17007/) (2025)

## Built on by (most cited first, 1 of 1)

- [Lightweight Polyp Segmentation via a Gain-Aware Prediction-Space Recursive Controller](https://mlsanity.com/paper/2607.03062/) (2026)

## Abstract (arXiv)

U-shaped networks output logits at multiple spatial scales, each capturing a different blend of coarse context and fine detail. Yet, training still treats these logits in isolation - either supervising only the final, highest-resolution logits or applying deep supervision with identical loss weights at every scale - without exploring mixed-scale combinations. Consequently, the decoder output misses the complementary cues that arise only when coarse and fine predictions are fused. To address this issue, we introduce LoMix (Logits Mixing), a NAS-inspired, differentiable plug-and-play module that generates new mixed-scale outputs and learns how exactly each of them should guide the training process. More precisely, LoMix mixes the multi-scale decoder logits with four lightweight fusion operators: addition, multiplication, concatenation, and attention-based weighted fusion, yielding a rich set of synthetic mutant maps. Every original or mutant map is given a softplus loss weight that is co-optimized with network parameters, mimicking a one-step architecture search that automatically discovers the most useful scales, mixtures, and operators. Plugging LoMix into recent U-shaped architectures (i.e., PVT-V2-B2 backbone with EMCAD decoder) on Synapse 8-organ dataset improves DICE by +4.2% over single-output supervision, +2.2% over deep supervision, and +1.5% over equally weighted additive fusion, all with zero inference overhead. When training data are scarce (e.g., one or two labeled scans), the advantage grows to +9.23%, underscoring LoMix's data efficiency. Across four benchmarks and diverse U-shaped networks, LoMiX improves DICE by up to +13.5% over single-output supervision, confirming that learnable weighted mixed-scale fusion generalizes broadly while remaining data efficient, fully interpretable, and overhead-free at inference. Our code is available at https://github.com/SLDGroup/LoMix.