# RePercENT: Scaling Disentangled Representation Learning Beyond Two Modalities

Source: [RePercENT: Scaling Disentangled Representation Learning Beyond Two Modalities](<https://www.sota2.com/research/paper/6a2106eec10484bac23f0669>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-05T05:00:12.194Z

Authors: Vasiliki Rizou, Pascal Frossard, Dorina Thanou

Published: 2026-06-03T17:10:39.000Z

## Abstract

To leverage the full potential of multimodal data, we need representations that go beyond the state-of-the-art alignment and fusion approaches and exploit all cross-modal interactions without sacrificing modality-specific information. Learning disentangled representations is a principled way to identify these underlying shared and unique factors that are hidden in observational data. However, while multimodal disentanglement is a compelling paradigm, existing methods are largely confined to the two-modality regime due to its inherent scalability bottleneck. To address this, we propose RePercENT, a self-supervised framework designed to surpass these limitations and unlocks scalable pairwise disentanglement beyond two modalities. Through a multimodal \`plug-and-play' architecture, our approach operates directly on pre-extracted embeddings, eliminating the need for extensive joint pre-training while making no assumptions regarding the underlying modalities or foundation model backbones. Moreover, we introduce a joint optimization objective for simultaneously deriving the shared and unique components, and provide formal theoretical guarantees that characterize the optimality of our solution. Across diverse modalities and tasks, RePercENT successfully recovers disentangled components while maintaining competitive performance and significantly reducing computational complexity.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Cancer Type Prediction on TCGA](<https://www.sota2.com/research/sota/cancer-type-prediction-on-tcga.md>): Cancer type prediction; TCGA; Accuracy (BRCA): 81.6.

- [Figurative Language Detection on IRFL](<https://www.sota2.com/research/sota/figurative-language-detection-on-irfl.md>): Figurative Language Detection; IRFL; Idioms Accuracy: 42.5.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.05109>)

[PDF](<https://arxiv.org/pdf/2606.05109.pdf>)
