# Disentangled and Interpretable Multimodal Attention Fusion for Cancer Survival Prediction

arXiv: https://arxiv.org/abs/2503.16069 · HTML: https://mlsanity.com/paper/2503.16069/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 32 references that are in the map.
- First posted on arXiv 20 Mar 2025.

## Builds on (most cited first, 14 of 14)

- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images](https://mlsanity.com/paper/2004.09666/) (2020)
- [Self-Normalizing Neural Networks](https://mlsanity.com/paper/1706.02515/) (2017)
- [CLUB: A Contrastive Log-ratio Upper Bound of Mutual Information](https://mlsanity.com/paper/2006.12013/) (2020)
- [Pathomic Fusion: An Integrated Framework for Fusing Histopathology and Genomic Features for Cancer Diagnosis and Prognosis](https://mlsanity.com/paper/1912.08937/) (2019)
- [Modeling Dense Multimodal Interactions Between Biological Pathways and Histology for Survival Prediction](https://mlsanity.com/paper/2304.06819/) (2023)
- [Multimodal Optimal Transport-based Co-Attention Transformer with Global Structure Consistency for Survival Prediction](https://mlsanity.com/paper/2306.08330/) (2023)
- [Morphological Prototyping for Unsupervised Slide Representation Learning in Computational Pathology](https://mlsanity.com/paper/2405.11643/) (2024)
- [Measuring Disentanglement: A Review of Metrics](https://mlsanity.com/paper/2012.09276/) (2020)
- [Multimodal Prototyping for cancer survival prediction](https://mlsanity.com/paper/2407.00224/) (2024)
- [Prototypical Information Bottlenecking and Disentangling for Multimodal Cancer Survival Prediction](https://mlsanity.com/paper/2401.01646/) (2024)
- [Measuring the Biases and Effectiveness of Content-Style Disentanglement](https://mlsanity.com/paper/2008.12378/) (2020)
- [DRIM: Learning Disentangled Representations from Incomplete Multimodal Healthcare Data](https://mlsanity.com/paper/2409.17055/) (2024)

## Built on by (most cited first, 4 of 4)

- [Beyond Attention Heatmaps: How to Get Better Explanations for Multiple Instance Learning Models in Histopathology](https://mlsanity.com/paper/2603.08328/) (2026)
- [Automated and Interpretable Survival Analysis from Multimodal Data](https://mlsanity.com/paper/2509.21600/) (2025)
- [Bridging the gap between Performance and Interpretability: An Explainable Disentangled Multimodal Framework for Cancer Survival Prediction](https://mlsanity.com/paper/2603.02162/) (2026)
- [RePercENT: Scaling Disentangled Representation Learning Beyond Two Modalities](https://mlsanity.com/paper/2606.05109/) (2026)

## Abstract (arXiv)

To improve the prediction of cancer survival using whole-slide images and transcriptomics data, it is crucial to capture both modality-shared and modality-specific information. However, multimodal frameworks often entangle these representations, limiting interpretability and potentially suppressing discriminative features. To address this, we propose Disentangled and Interpretable Multimodal Attention Fusion (DIMAF), a multimodal framework that separates the intra- and inter-modal interactions within an attention-based fusion mechanism to learn distinct modality-specific and modality-shared representations. We introduce a loss based on Distance Correlation to promote disentanglement between these representations and integrate Shapley additive explanations to assess their relative contributions to survival prediction. We evaluate DIMAF on four public cancer survival datasets, achieving a relative average improvement of 1.85% in performance and 23.7% in disentanglement compared to current state-of-the-art multimodal models. Beyond improved performance, our interpretable framework enables a deeper exploration of the underlying interactions between and within modalities in cancer biology.