Disentanglement of Variations with Multimodal Generative Modeling

• URL: http://arxiv.org/abs/2509.23548v1
• Date: Sun, 28 Sep 2025 00:54:39 GMT
• Title: Disentanglement of Variations with Multimodal Generative Modeling
• Authors: Yijie Zhang, Yiyang Shen, Weiran Wang,
• Abstract summary: We propose Information-disentangled Multimodal VAE (IDMVAE) to address the issue of learning robust representations of multimodal data.<n>Compared to existing approaches, IDMVAE shows a clean separation between shared and private information, demonstrating superior generation quality and semantic coherence on challenging datasets.
• Score: 18.610989859219895
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal data are prevalent across various domains, and learning robust representations of such data is paramount to enhancing generation quality and downstream task performance. To handle heterogeneity and interconnections among different modalities, recent multimodal generative models extract shared and private (modality-specific) information with two separate variables. Despite attempts to enforce disentanglement between these two variables, these methods struggle with challenging datasets where the likelihood model is insufficient. In this paper, we propose Information-disentangled Multimodal VAE (IDMVAE) to explicitly address this issue, with rigorous mutual information-based regularizations, including cross-view mutual information maximization for extracting shared variables, and a cycle-consistency style loss for redundancy removal using generative augmentations. We further introduce diffusion models to improve the capacity of latent priors. These newly proposed components are complementary to each other. Compared to existing approaches, IDMVAE shows a clean separation between shared and private information, demonstrating superior generation quality and semantic coherence on challenging datasets.

Related papers

• MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings [75.0617088717528]
  MoCa is a framework for transforming pre-trained VLM backbones into effective bidirectional embedding models.<n>MoCa consistently improves performance across MMEB and ViDoRe-v2 benchmarks, achieving new state-of-the-art results.
  arXiv  Detail & Related papers  (2025-06-29T06:41:00Z)
• Learning Multimodal Latent Generative Models with Energy-Based Prior [3.6648642834198797]
  We propose a novel framework that integrates the latent generative model with the EBM. This approach results in a more expressive and informative prior, better-capturing of information across multiple modalities.
  arXiv  Detail & Related papers  (2024-09-30T01:38:26Z)
• Leveraging Diffusion Disentangled Representations to Mitigate Shortcuts in Underspecified Visual Tasks [92.32670915472099]
  We propose an ensemble diversification framework exploiting the generation of synthetic counterfactuals using Diffusion Probabilistic Models (DPMs) We show that diffusion-guided diversification can lead models to avert attention from shortcut cues, achieving ensemble diversity performance comparable to previous methods requiring additional data collection.
  arXiv  Detail & Related papers  (2023-10-03T17:37:52Z)
• Learning multi-modal generative models with permutation-invariant encoders and tighter variational objectives [5.549794481031468]
  Devising deep latent variable models for multi-modal data has been a long-standing theme in machine learning research. In this work, we consider a variational objective that can tightly approximate the data log-likelihood. We develop more flexible aggregation schemes that avoid the inductive biases in PoE or MoE approaches.
  arXiv  Detail & Related papers  (2023-09-01T10:32:21Z)
• Improving Multimodal Joint Variational Autoencoders through Normalizing Flows and Correlation Analysis [0.0]
  The unimodal posteriors are conditioned on the Deep Canonical Correlation Analysis embeddings. We also use Normalizing Flows to enrich the unimodal posteriors and achieve more diverse data generation.
  arXiv  Detail & Related papers  (2023-05-19T17:15:34Z)
• Align and Attend: Multimodal Summarization with Dual Contrastive Losses [57.83012574678091]
  The goal of multimodal summarization is to extract the most important information from different modalities to form output summaries. Existing methods fail to leverage the temporal correspondence between different modalities and ignore the intrinsic correlation between different samples. We introduce Align and Attend Multimodal Summarization (A2Summ), a unified multimodal transformer-based model which can effectively align and attend the multimodal input.
  arXiv  Detail & Related papers  (2023-03-13T17:01:42Z)
• Discriminative Multimodal Learning via Conditional Priors in Generative Models [21.166519800652047]
  This research studies the realistic scenario in which all modalities and class labels are available for model training. We show, in this scenario, that the variational lower bound limits mutual information between joint representations and missing modalities.
  arXiv  Detail & Related papers  (2021-10-09T17:22:24Z)
• Bi-Bimodal Modality Fusion for Correlation-Controlled Multimodal Sentiment Analysis [96.46952672172021]
  Bi-Bimodal Fusion Network (BBFN) is a novel end-to-end network that performs fusion on pairwise modality representations. Model takes two bimodal pairs as input due to known information imbalance among modalities.
  arXiv  Detail & Related papers  (2021-07-28T23:33:42Z)
• Learning Multimodal VAEs through Mutual Supervision [72.77685889312889]
  MEME combines information between modalities implicitly through mutual supervision. We demonstrate that MEME outperforms baselines on standard metrics across both partial and complete observation schemes.
  arXiv  Detail & Related papers  (2021-06-23T17:54:35Z)
• Private-Shared Disentangled Multimodal VAE for Learning of Hybrid Latent Representations [24.3033562693679]
  We introduce a disentangled multi-modal variational autoencoder (DMVAE) that utilizes disentangled VAE strategy to separate the private and shared latent spaces of multiple modalities. We demonstrate the utility of DMVAE on a semi-supervised learning task, where one of the modalities contains partial data labels. Our experiments on several benchmarks indicate the importance of the private-shared disentanglement as well as the hybrid latent representation.
  arXiv  Detail & Related papers  (2020-12-23T23:33:23Z)
• Relating by Contrasting: A Data-efficient Framework for Multimodal Generative Models [86.9292779620645]
  We develop a contrastive framework for generative model learning, allowing us to train the model not just by the commonality between modalities, but by the distinction between "related" and "unrelated" multimodal data. Under our proposed framework, the generative model can accurately identify related samples from unrelated ones, making it possible to make use of the plentiful unlabeled, unpaired multimodal data.
  arXiv  Detail & Related papers  (2020-07-02T15:08:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.