# SLIM-Diff: Shared Latent Image-Mask Diffusion with Lp loss for Data-Scarce Epilepsy FLAIR MRI

arXiv: https://arxiv.org/abs/2602.03372 · HTML: https://mlsanity.com/paper/2602.03372/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models.
- Builds on 8 of its 18 references that are in the map.
- First posted on arXiv 3 Feb 2026.

## Placed in

- Diffusion models / [Medical imaging and segmentation](https://mlsanity.com/diffusion/medical-imaging-and-segmentation/)

## Builds on (most cited first, 8 of 8)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Demystifying MMD GANs](https://mlsanity.com/paper/1801.01401/) (2018)
- [A General and Adaptive Robust Loss Function](https://mlsanity.com/paper/1701.03077/) (2017)
- [Diffusion Probabilistic Models beat GANs on Medical Images](https://mlsanity.com/paper/2212.07501/) (2022)
- [MedSegFactory: Text-Guided Generation of Medical Image-Mask Pairs](https://mlsanity.com/paper/2504.06897/) (2025)

## Abstract (arXiv)

Focal cortical dysplasia (FCD) lesions in epilepsy FLAIR MRI are subtle and scarce, making joint image--mask generative modeling prone to instability and memorization. We propose SLIM-Diff, a compact joint diffusion model whose main contributions are (i) a single shared-bottleneck U-Net that enforces tight coupling between anatomy and lesion geometry from a 2-channel image+mask representation, and (ii) loss-geometry tuning via a tunable Lp objective. As an internal baseline, we include the canonical DDPM-style objective (ε-prediction with L2 loss) and isolate the effect of prediction parameterization and Lp geometry under a matched setup. Experiments show that x0-prediction is consistently the strongest choice for joint synthesis, and that fractional sub-quadratic penalties (L1.5) improve image fidelity while L2 better preserves lesion mask morphology. Our code and model weights are available in https://github.com/MarioPasc/slim-diff