Salvato in:
Dettagli Bibliografici
Autori principali: Role, François, Meyer, Sébastien, Amblard, Victor
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2505.03703
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913823227641856
author Role, François
Meyer, Sébastien
Amblard, Victor
author_facet Role, François
Meyer, Sébastien
Amblard, Victor
contents Vision-language models (VLMs) allow to embed texts and images in a shared representation space. However, it has been shown that these models are subject to a modality gap phenomenon meaning there exists a clear separation between the embeddings from one modality and another in the embedding space. While this misalignment is detrimental for downstream tasks such as multimodal retrieval, multimodal clustering or zero-shot classification, etc. no generic and practical methods have so far been proposed to assess it precisely and even reduce it. We therefore propose novel measures and effective techniques (spectral- and optimal transport-based methods) to achieve this goal. Extensive experiments conducted on several image-text datasets and models demonstrate their effectiveness and beneficial effects on downstream tasks. Our code is available at the URL provided in the paper's abstract.
format Preprint
id arxiv_https___arxiv_org_abs_2505_03703
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Fill the Gap: Quantifying and Reducing the Modality Gap in Image-Text Representation Learning
Role, François
Meyer, Sébastien
Amblard, Victor
Computer Vision and Pattern Recognition
Machine Learning
Vision-language models (VLMs) allow to embed texts and images in a shared representation space. However, it has been shown that these models are subject to a modality gap phenomenon meaning there exists a clear separation between the embeddings from one modality and another in the embedding space. While this misalignment is detrimental for downstream tasks such as multimodal retrieval, multimodal clustering or zero-shot classification, etc. no generic and practical methods have so far been proposed to assess it precisely and even reduce it. We therefore propose novel measures and effective techniques (spectral- and optimal transport-based methods) to achieve this goal. Extensive experiments conducted on several image-text datasets and models demonstrate their effectiveness and beneficial effects on downstream tasks. Our code is available at the URL provided in the paper's abstract.
title Fill the Gap: Quantifying and Reducing the Modality Gap in Image-Text Representation Learning
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2505.03703