Enregistré dans:
Détails bibliographiques
Auteurs principaux: Sen, Souptik, Younis, Raneen, Ahmadi, Zahra
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2605.12145
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909039074476032
author Sen, Souptik
Younis, Raneen
Ahmadi, Zahra
author_facet Sen, Souptik
Younis, Raneen
Ahmadi, Zahra
contents Multimodal learning seeks to integrate information across diverse sensory sources, yet current approaches struggle to balance cross-modal generalizability with modality-specific structure. Continuous (implicit) methods preserve fine-grained priors but render generalization challenging, while discrete (explicit) approaches enforce shared prototypes at the expense of modality specificity. We introduce CoDAAR (Cross-modal Discrete Alignment And Reconstruction), a novel framework that resolves this long-standing trade-off by establishing semantic consensus across modality-specific codebooks through index-level alignment. This design uniquely allows CoDAAR to preserve modality-unique structures while achieving generalizable cross-modal representations within a unified discrete space. CoDAAR combines two complementary mechanisms: Discrete Temporal Alignment (DTA), which enables fine-grained temporal quantization, and Cascading Semantic Alignment (CSA), which promotes progressive cross-modal semantic agreement. Together, they establish a competition-free unified representation space. Trained with self-supervised reconstruction objectives on paired multimodal sequences, CoDAAR demonstrates robust cross-modal and cross-domain generalization. Across Cross-Modal Generalization benchmarks, including event classification, localization, video segmentation, and cross-dataset transfer, CoDAAR achieves state-of-the-art performance, establishing a new paradigm for discrete and generalizable multimodal representation learning.
format Preprint
id arxiv_https___arxiv_org_abs_2605_12145
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Cross-Modal-Domain Generalization Through Semantically Aligned Discrete Representations
Sen, Souptik
Younis, Raneen
Ahmadi, Zahra
Computer Vision and Pattern Recognition
Multimodal learning seeks to integrate information across diverse sensory sources, yet current approaches struggle to balance cross-modal generalizability with modality-specific structure. Continuous (implicit) methods preserve fine-grained priors but render generalization challenging, while discrete (explicit) approaches enforce shared prototypes at the expense of modality specificity. We introduce CoDAAR (Cross-modal Discrete Alignment And Reconstruction), a novel framework that resolves this long-standing trade-off by establishing semantic consensus across modality-specific codebooks through index-level alignment. This design uniquely allows CoDAAR to preserve modality-unique structures while achieving generalizable cross-modal representations within a unified discrete space. CoDAAR combines two complementary mechanisms: Discrete Temporal Alignment (DTA), which enables fine-grained temporal quantization, and Cascading Semantic Alignment (CSA), which promotes progressive cross-modal semantic agreement. Together, they establish a competition-free unified representation space. Trained with self-supervised reconstruction objectives on paired multimodal sequences, CoDAAR demonstrates robust cross-modal and cross-domain generalization. Across Cross-Modal Generalization benchmarks, including event classification, localization, video segmentation, and cross-dataset transfer, CoDAAR achieves state-of-the-art performance, establishing a new paradigm for discrete and generalizable multimodal representation learning.
title Cross-Modal-Domain Generalization Through Semantically Aligned Discrete Representations
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.12145