Saved in:
Bibliographic Details
Main Authors: Kritikos, Antonios, Spanos, Nikolaos, Voulodimos, Athanasios
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.16892
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908975964880896
author Kritikos, Antonios
Spanos, Nikolaos
Voulodimos, Athanasios
author_facet Kritikos, Antonios
Spanos, Nikolaos
Voulodimos, Athanasios
contents Domain generalization (DG) aims to maintain performance under domain shift, which in computer vision appears primarily as stylistic variations that cause models to overfit to domain-specific appearance cues rather than class semantics. To overcome this, recent methods use textual representations as stable, domain-invariant anchors. However, multimodal approaches that rely on cosine similarity-based contrastive alignment leave a modality gap where image and text embeddings remain geometrically separated despite semantic correspondence. We propose CrossFlowDG, a novel DG framework that addresses this residual gap using noise-free, cross-modal flow matching. By learning a continuous transformation in the joint Euclidean latent space, our framework explicitly transports domain-biased image embeddings toward domain-invariant text embeddings of the correct class. Using the efficient VMamba image encoder and CLIP's text encoder, CrossFlowDG is tested against four common DG benchmarks, and achieves competitive performance on several benchmarks and state-of-the-art on TerraIncognita. Code is available at: https://github.com/ajkrit/CrossFlowDG
format Preprint
id arxiv_https___arxiv_org_abs_2604_16892
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CrossFlowDG: Bridging the Modality Gap with Cross-modal Flow Matching for Domain Generalization
Kritikos, Antonios
Spanos, Nikolaos
Voulodimos, Athanasios
Computer Vision and Pattern Recognition
Domain generalization (DG) aims to maintain performance under domain shift, which in computer vision appears primarily as stylistic variations that cause models to overfit to domain-specific appearance cues rather than class semantics. To overcome this, recent methods use textual representations as stable, domain-invariant anchors. However, multimodal approaches that rely on cosine similarity-based contrastive alignment leave a modality gap where image and text embeddings remain geometrically separated despite semantic correspondence. We propose CrossFlowDG, a novel DG framework that addresses this residual gap using noise-free, cross-modal flow matching. By learning a continuous transformation in the joint Euclidean latent space, our framework explicitly transports domain-biased image embeddings toward domain-invariant text embeddings of the correct class. Using the efficient VMamba image encoder and CLIP's text encoder, CrossFlowDG is tested against four common DG benchmarks, and achieves competitive performance on several benchmarks and state-of-the-art on TerraIncognita. Code is available at: https://github.com/ajkrit/CrossFlowDG
title CrossFlowDG: Bridging the Modality Gap with Cross-modal Flow Matching for Domain Generalization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.16892