English
Related papers

Related papers: Canonicalizing Multimodal Contrastive Representati…

200 papers

Multi-modal contrastive learning as a self-supervised representation learning technique has achieved great success in foundation model training, such as CLIP~\citep{radford2021learning}. In this paper, we study the theoretical properties of…

Machine Learning · Statistics 2025-05-20 Yu Gui , Cong Ma , Zongming Ma

The Platonic Representation Hypothesis posits that learned representations from models trained on different modalities converge to a shared latent structure of the world. However, this hypothesis has largely been examined in vision and…

Artificial Intelligence · Computer Science 2026-02-24 Pratham Yashwante , Rose Yu

Contrastive Language-Image Pre-training (CLIP) has shown impressive performance in aligning visual and textual representations. Recent studies have extended this paradigm to 3D vision to improve scene understanding for autonomous driving. A…

Computer Vision and Pattern Recognition · Computer Science 2026-03-10 Ximeng Tao , Dimitar Filev , Gaurav Pandey

Multi-modal semantic understanding requires integrating information from different modalities to extract users' real intention behind words. Most previous work applies a dual-encoder structure to separately encode image and text, but fails…

Computation and Language · Computer Science 2024-03-12 Ming Zhang , Ke Chang , Yunfang Wu

Image-text representation learning forms a cornerstone in vision-language models, where pairs of images and textual descriptions are contrastively aligned in a shared embedding space. Since visual and textual concepts are naturally…

Computer Vision and Pattern Recognition · Computer Science 2025-03-04 Avik Pal , Max van Spengler , Guido Maria D'Amely di Melendugno , Alessandro Flaborea , Fabio Galasso , Pascal Mettes

Contrastive learning-based vision-language pre-training approaches, such as CLIP, have demonstrated great success in many vision-language tasks. These methods achieve cross-modal alignment by encoding a matched image-text pair with similar…

Computer Vision and Pattern Recognition · Computer Science 2023-03-28 Yuxiao Chen , Jianbo Yuan , Yu Tian , Shijie Geng , Xinyu Li , Ding Zhou , Dimitris N. Metaxas , Hongxia Yang

Spatial understanding remains a key challenge in vision-language models. Yet it is still unclear whether such understanding is truly acquired, and if so, through what mechanisms. We present a controllable 1D image-text testbed to probe how…

Computer Vision and Pattern Recognition · Computer Science 2026-05-27 Takaki Yamamoto , Chihiro Noguchi , Toshihiro Tanizawa

Multimodal representation learning is fundamentally about transforming incomparable modalities into comparable representations. While prior research primarily focused on explicitly aligning these representations through targeted learning…

Machine Learning · Computer Science 2025-06-16 Megan Tjandrasuwita , Chanakya Ekbote , Liu Ziyin , Paul Pu Liang

We tackle the problem of learning the geometry of multiple categories of deformable objects jointly. Recent work has shown that it is possible to learn a unified dense pose predictor for several categories of related objects. However,…

Computer Vision and Pattern Recognition · Computer Science 2021-06-21 Natalia Neverova , Artsiom Sanakoyeu , Patrick Labatut , David Novotny , Andrea Vedaldi

Aligned text-image encoders such as CLIP have become the de facto model for vision-language tasks. Furthermore, modality-specific encoders achieve impressive performances in their respective domains. This raises a central question: does an…

Foundation machine learning interatomic potentials (MLIPs) are trained on overlapping chemical spaces, yet their latent representations remain model-specific. Here, we show that independently developed MLIPs exhibit statistically consistent…

Materials Science · Physics 2025-12-08 Zhenzhu Li , Aron Walsh

Cross-modal retrieval is the task of retrieving samples of a given modality by using queries of a different one. Due to the wide range of practical applications, the problem has been mainly focused on the vision and language case, e.g. text…

Computer Vision and Pattern Recognition · Computer Science 2024-01-30 Jorge Sánchez , Rodrigo Laguna

Multimodal federated learning in real-world settings often encounters incomplete and heterogeneous data across clients. This results in misaligned local feature representations that limit the effectiveness of model aggregation. Unlike prior…

Machine Learning · Computer Science 2025-10-28 Duong M. Nguyen , Trong Nghia Hoang , Thanh Trung Huynh , Quoc Viet Hung Nguyen , Phi Le Nguyen

The popularity of self-supervised learning has made it possible to train models without relying on labeled data, which saves expensive annotation costs. However, most existing self-supervised contrastive learning methods often overlook the…

Computer Vision and Pattern Recognition · Computer Science 2023-08-01 Weiquan Li , Xianzhong Long , Yun Li

In typical multimodal contrastive learning, such as CLIP, encoders produce one point in the latent representation space for each input. However, one-point representation has difficulty in capturing the relationship and the similarity…

Machine Learning · Computer Science 2025-03-04 Toshimitsu Uesaka , Taiji Suzuki , Yuhta Takida , Chieh-Hsin Lai , Naoki Murata , Yuki Mitsufuji

Machine learning models of vastly different modalities and architectures are being trained to predict the behavior of molecules, materials, and proteins. However, it remains unclear whether they learn similar internal representations of…

Machine Learning · Computer Science 2025-12-04 Sathya Edamadaka , Soojung Yang , Ju Li , Rafael Gómez-Bombarelli

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

Multimodal learning seeks to integrate information from heterogeneous sources, where signals may be shared across modalities, specific to individual modalities, or emerge only through their interaction. While self-supervised multimodal…

Machine Learning · Computer Science 2026-02-17 Carolin Cissee , Raneen Younis , Zahra Ahmadi

Large Multimodal Models (LMMs) typically build on ViTs (e.g., CLIP), yet their training with simple random in-batch negatives limits the ability to capture fine-grained visual differences, particularly in geometric scenarios. To address…

Computer Vision and Pattern Recognition · Computer Science 2025-10-02 Kai Sun , Yushi Bai , Zhen Yang , Jiajie Zhang , Ji Qi , Lei Hou , Juanzi Li

While InfoNCE underlies modern contrastive learning, its geometric mechanisms remain under-characterized beyond the canonical alignment--uniformity decomposition. We develop a measure-theoretic framework in which representation measures…

Machine Learning · Computer Science 2026-05-26 Yichao Cai , Zhen Zhang , Yuhang Liu , Javen Qinfeng Shi