# SNR-Adaptive Unified Diffusion for Multi-Task Medical Image Segmentation

arXiv: https://arxiv.org/abs/2607.03103 · HTML: https://mlsanity.com/paper/2607.03103/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 28 references that are in the map.
- First posted on arXiv 3 Jul 2026.

## Builds on (most cited first, 16 of 16)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [Uncertainty-aware Self-ensembling Model for Semi-supervised 3D Left Atrium Segmentation](https://mlsanity.com/paper/1907.07034/) (2019)
- [MedSegDiff: Medical Image Segmentation with Diffusion Probabilistic Model](https://mlsanity.com/paper/2211.00611/) (2022)
- [Unsupervised Bidirectional Cross-Modality Adaptation via Deeply Synergistic Image and Feature Alignment for Medical Image Segmentation](https://mlsanity.com/paper/2002.02255/) (2020)
- [Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domains](https://mlsanity.com/paper/2007.02035/) (2020)
- [Bidirectional Copy-Paste for Semi-Supervised Medical Image Segmentation](https://mlsanity.com/paper/2305.00673/) (2023)
- [Semi-supervised Left Atrium Segmentation with Mutual Consistency Training](https://mlsanity.com/paper/2103.02911/) (2021)
- [Exploring Smoothness and Class-Separation for Semi-supervised Medical Image Segmentation](https://mlsanity.com/paper/2203.01324/) (2022)
- [Momentum Contrastive Voxel-wise Representation Learning for Semi-supervised Volumetric Medical Image Segmentation](https://mlsanity.com/paper/2105.07059/) (2021)
- [Deep Symmetric Adaptation Network for Cross-modality Medical Image Segmentation](https://mlsanity.com/paper/2101.06853/) (2021)
- [Semi-supervised Meta-learning with Disentanglement for Domain-generalised Medical Image Segmentation](https://mlsanity.com/paper/2106.13292/) (2021)
- [Towards Generic Semi-Supervised Framework for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/2310.11320/) (2023)
- [vMFNet: Compositionality Meets Domain-generalised Segmentation](https://mlsanity.com/paper/2206.14538/) (2022)
- [Diffuse-UDA: Addressing Unsupervised Domain Adaptation in Medical Image Segmentation with Appearance and Structure Aligned Diffusion Models](https://mlsanity.com/paper/2408.05985/) (2024)

## Abstract (arXiv)

Clinical cardiac imaging pipelines currently deploy separate models for each dataset and modality, incurring redundant training costs and precluding knowledge sharing across anatomically related tasks. Consolidating semi-supervised learning, unsupervised domain adaptation, and domain generalisation into one model is therefore a practical necessity, yet naive joint training exposes a fundamental barrier: conflicting label semantics between datasets collapse LA Dice from 90.31% to 83.38%, while gradient imbalance across tasks of unequal complexity suppresses the weaker tasks throughout training. We present UniT-Diff, a unified diffusion segmentation framework that resolves these conflicts through three targeted mechanisms. An 11-channel task-specific output space physically partitions label categories, eliminating cross-task gradient sign reversal by construction. SNR-Adaptive Task Conditioning (SATC) scales the task token by the log signal-to-noise ratio of the current diffusion timestep, suppressing domain-specific bias during coarse denoising and restoring full task guidance as the signal clears. Task-Type-Aware Conditional Dropout (TTACD) permanently removes the task token for domain-generalisation inputs, routing them through a shared neutral pathway that draws on cross-dataset cardiac anatomy rather than source-vendor statistics. Under a single parameter set, UniT-Diff surpasses independently trained task-specific baselines on all three benchmarks simultaneously: +0.87% on LA, +1.77% on MMWHS, and +0.88% on MNMS.