English
Related papers

Related papers: Bridging the Modality Gap: Dimension Information A…

200 papers

The success of vision-language models is primarily attributed to effective alignment across modalities such as vision and language. However, modality gaps persist in existing alignment algorithms and appear necessary for human perception as…

Computer Vision and Pattern Recognition · Computer Science 2026-04-28 Hanqi Yan , Xiangxiang Cui , Lu Yin , Jindong Gu , Paul Pu Liang , Yulan He , Yifei Wang

We present modality gap, an intriguing geometric phenomenon of the representation space of multi-modal models. Specifically, we show that different data modalities (e.g. images and text) are embedded at arm's length in their shared…

Computation and Language · Computer Science 2022-10-21 Weixin Liang , Yuhui Zhang , Yongchan Kwon , Serena Yeung , James Zou

Multi-modal contrastive models such as CLIP achieve state-of-the-art performance in zero-shot classification by embedding input images and texts on a joint representational space. Recently, a modality gap has been reported in two-encoder…

Computer Vision and Pattern Recognition · Computer Science 2024-06-10 Abrar Fahim , Alex Murphy , Alona Fyshe

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…

Computer Vision and Pattern Recognition · Computer Science 2025-05-07 François Role , Sébastien Meyer , Victor Amblard

Despite the achievements of large-scale multimodal pre-training approaches, cross-modal retrieval, e.g., image-text retrieval, remains a challenging task. To bridge the semantic gap between the two modalities, previous studies mainly focus…

Computation and Language · Computer Science 2021-06-08 Shuhuai Ren , Junyang Lin , Guangxiang Zhao , Rui Men , An Yang , Jingren Zhou , Xu Sun , Hongxia Yang

Many modern multi-modal models (e.g. CLIP) seek an embedding space in which the two modalities are aligned. Somewhat surprisingly, almost all existing models show a strong modality gap: the distribution of images is well-separated from the…

Computer Vision and Pattern Recognition · Computer Science 2026-04-29 Rhea Chowers , Oshri Naparstek , Udi Barzelay , Yair Weiss

Cross-modal alignment is a crucial task in multimodal learning aimed at achieving semantic consistency between vision and language. This requires that image-text pairs exhibit similar semantics. Traditional algorithms pursue embedding…

Machine Learning · Computer Science 2026-03-09 Xiang Ma , Lexin Fang , Litian Xu , Caiming Zhang

Although image captioning models have made significant advancements in recent years, the majority of them heavily depend on high-quality datasets containing paired images and texts which are costly to acquire. Previous works leverage the…

Computer Vision and Pattern Recognition · Computer Science 2023-12-15 Zhiyue Liu , Jinyuan Liu , Fanrong Ma

In multimedia applications, the text and image components in a web document form a pairwise constraint that potentially indicates the same semantic concept. This paper studies cross-modal learning via the pairwise constraint, and aims to…

Computer Vision and Pattern Recognition · Computer Science 2023-07-19 Ran He , Man Zhang , Liang Wang , Ye Ji , Qiyue Yin

Despite the success of multimodal contrastive learning in aligning visual and linguistic representations, a persistent geometric anomaly, the Modality Gap, remains: embeddings of distinct modalities expressing identical semantics occupy…

Computer Vision and Pattern Recognition · Computer Science 2026-05-11 Xiaomin Yu , Yi Xin , Yuhui Zhang , Wenjie Zhang , Chonghan Liu , Hanzhen Zhao , Chen Liu , Xiaoxing Hu , Ziyue Qiao , Hao Tang , Xiaobin Hu , Chengwei Qin , Hui Xiong , Yu Qiao , Shuicheng Yan

In this paper, we study the cross-modal image retrieval, where the inputs contain a source image plus some text that describes certain modifications to this image and the desired image. Prior work usually uses a three-stage strategy to…

Computer Vision and Pattern Recognition · Computer Science 2021-03-11 Chunbin Gu , Jiajun Bu , Xixi Zhou , Chengwei Yao , Dongfang Ma , Zhi Yu , Xifeng Yan

Contrastive Language--Image Pre-training (CLIP) has manifested remarkable improvements in zero-shot classification and cross-modal vision-language tasks. Yet, from a geometrical point of view, the CLIP embedding space has been found to have…

Computer Vision and Pattern Recognition · Computer Science 2024-09-17 Sedigheh Eslami , Gerard de Melo

Text-image cross-modal retrieval is a challenging task in the field of language and vision. Most previous approaches independently embed images and sentences into a joint embedding space and compare their similarities. However, previous…

Computer Vision and Pattern Recognition · Computer Science 2019-09-13 Zihao Wang , Xihui Liu , Hongsheng Li , Lu Sheng , Junjie Yan , Xiaogang Wang , Jing Shao

Human perception integrates multiple modalities, such as vision, hearing, and language, into a unified understanding of the surrounding reality. While recent multimodal models have achieved significant progress by aligning pairs of…

Computer Vision and Pattern Recognition · Computer Science 2025-02-13 Giordano Cicchetti , Eleonora Grassucci , Luigi Sigillo , Danilo Comminiello

Multimodal learning has recently gained significant popularity, demonstrating impressive performance across various zero-shot classification tasks and a range of perceptive and generative applications. Models such as Contrastive…

Machine Learning · Computer Science 2026-02-16 Can Yaras , Siyi Chen , Peng Wang , Qing Qu

Cross-modal alignment aims to map heterogeneous modalities into a shared latent space, as exemplified by models like CLIP, which benefit from large-scale image-text pretraining for strong recognition capabilities. However, when operating in…

Computer Vision and Pattern Recognition · Computer Science 2025-10-27 Jiaxiang Liu , Yuan Wang , Jiawei Du , Joey Tianyi Zhou , Mingkun Xu , Zuozhu Liu

Advances in vision-language models (VLMs) have enabled effective cross-modality retrieval. However, when both text and images exist in the database, similarity scores would differ in scale by modality. This phenomenon, known as the modality…

Computation and Language · Computer Science 2025-12-01 Shuhei Yamashita , Daiki Shirafuji , Tatsuhiko Saito

Vision-Language Models (VLMs) such as CLIP learn a shared embedding space for images and text, yet their representations remain geometrically separated, a phenomenon known as the modality gap. This gap limits tasks requiring cross-modal…

Computer Vision and Pattern Recognition · Computer Science 2026-04-02 Hongyuan Liu , Qinli Yang , Wen Li , Zhong Zhang , Jiaming Liu , Wei Han , Zhili Qin , Jinxia Guo , Junming Shao

In multimodal learning, CLIP has emerged as the de-facto approach for mapping different modalities into a shared latent space by bringing semantically similar representations closer while pushing apart dissimilar ones. However, CLIP-based…

Computer Vision and Pattern Recognition · Computer Science 2026-02-24 Eleonora Grassucci , Giordano Cicchetti , Danilo Comminiello

Multimodal models often over-rely on dominant modalities, failing to achieve optimal performance. While prior work focuses on modifying training objectives or optimization procedures, data-centric solutions remain underexplored. We propose…

Machine Learning · Computer Science 2025-10-01 Seong-Hyeon Hwang , Soyoung Choi , Steven Euijong Whang
‹ Prev 1 2 3 10 Next ›