# Synergistic Information Disentanglement for Omni-modal Slide Representation Learning in Computational Pathology

arXiv: https://arxiv.org/abs/2609.02118 · HTML: https://mlsanity.com/paper/2609.02118/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 29 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 18 of 18)

- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images](https://mlsanity.com/paper/2004.09666/) (2020)
- [GatorTron: A Large Clinical Language Model to Unlock Patient Information from Unstructured Electronic Health Records](https://mlsanity.com/paper/2203.03540/) (2022)
- [Accelerating Data Processing and Benchmarking of AI Models for Pathology](https://mlsanity.com/paper/2502.06750/) (2025)
- [A Multimodal Knowledge-enhanced Whole-slide Pathology Foundation Model](https://mlsanity.com/paper/2407.15362/) (2024)
- [Transcriptomics-guided Slide Representation Learning in Computational Pathology](https://mlsanity.com/paper/2405.11618/) (2024)
- [Multistain Pretraining for Slide Representation Learning in Pathology](https://mlsanity.com/paper/2408.02859/) (2024)
- [Gaussian Partial Information Decomposition: Bias Correction and Application to High-dimensional Data](https://mlsanity.com/paper/2307.10515/) (2023)
- [Unsupervised Foundation Model-Agnostic Slide-Level Representation Learning](https://mlsanity.com/paper/2411.13623/) (2024)
- [Disentanglement Analysis with Partial Information Decomposition](https://mlsanity.com/paper/2108.13753/) (2021)
- [MGCT: Mutual-Guided Cross-Modality Transformer for Survival Outcome Prediction using Integrative Histopathology-Genomic Features](https://mlsanity.com/paper/2311.11659/) (2023)
- [HoneyBee: A Scalable Modular Framework for Creating Multimodal Oncology Datasets with Foundational Embedding Models](https://mlsanity.com/paper/2405.07460/) (2024)
- [Unleashing the Infinity Power of Geometry: A Novel Geometry-Aware Transformer (GOAT) for Whole Slide Histopathology Image Analysis](https://mlsanity.com/paper/2402.05373/) (2024)
- [MIRROR: Multi-Modal Pathological Self-Supervised Representation Learning via Modality Alignment and Retention](https://mlsanity.com/paper/2503.00374/) (2025)
- [MurreNet: Modeling Holistic Multimodal Interactions Between Histopathology and Genomic Profiles for Survival Prediction](https://mlsanity.com/paper/2507.04891/) (2025)
- [Disentangled Representations via Synergy Minimization](https://mlsanity.com/paper/1710.03839/) (2017)
- [BioX-CPath: Biologically-driven Explainable Diagnostics for Multistain IHC Computational Pathology](https://mlsanity.com/paper/2503.20880/) (2025)

## Abstract (arXiv)

In computational pathology (CPath), developing omni-modal self-supervised learning (SSL) models that integrate histology, genomics, and clinical reports enables transferable representation learning for whole slide images (WSIs). Existing approaches implicitly force heterogeneous modalities into a uniform latent space by contrastive alignment, causing modality collapse where unique, synergistic diagnostic signals (termed as Phi) are discarded in favor of trivial redundancy. We hypothesize that the strongest task-agnostic SSL training signal stems from distilling the synergistic interactions over merely aligning shared redundancy. To this end, we introduce textscPhi-Omni, a synergistic information disentanglement framework grounded in Partial Information Decomposition (PID) theory for slide representation learning. Unlike standard contrastive approaches, textscPhi-Omni employs a Synergistic Information Bottleneck (SIB) regulated by the proposed PhiID objective, which explicitly suppresses marginal redundancy while maximizing irreducible synergy, thereby distilling high-order cross-modal interactions. Following pretraining on breast (n=1031) and lung (n=919) cohorts, textscPhi-Omni demonstrates superior few-shot performance across five independent external datasets spanning eight tasks compared to supervised and SSL baselines. Source code is available here.