CORTEXA
← Browse
arxivcs.CV2026-07-08

A Theory of Contrastive Learning with Natural Images

Antonio Torralba, Yair Weiss

Why does contrastive learning with simple images and augmentations yield useful representations for downstream tasks? We address this question by analytically computing the optimal representation in terms of a contrastive loss for a range of basic augmentations and any image dataset with stationary statistics. We show that for certain augmentations the optimum can be attained by a CNN whose first layer filters are sinusoids, followed by a pointwise nonlinearity, global average pooling, and a final linear layer that performs partial whitening. We also show that the optimal weights in such CNNs for more complicated augmentations are still sinusoids. The frequencies of the sinusoids and their weights can be computed using a simple waterfilling algorithm given the dataset's expected power spectrum. Experiments with different image datasets and augmentations show that such CNNs trained with SGD empirically learn sinusoids in their first layer and to perform partial whitening

View free PDFSource page

Related papers

arxivcs.CV2026-07-23

Webly Supervised Multi-Label Recognition: Evaluation Benchmark and Dual-Branch Multi-Label Contrastive Learning

Zhihua Xu, Zhijing Yang, Yufeng Yang, Tianshui Chen

Training deep learning models with freely available web images can reduce their dependence on costly manual annotations. Although webly supervised learning has been widely studied for single-label recognition, its multi-label counterpart remains underexplored, partly due to the l…

View free PDFSource page
arxivcs.CV2026-07-09

Vision Transformers Learn Gestalt-Like Figure-Ground Cues from Natural Images

Matthias Tangemann, Benjamin Lo, Zygmunt Pizlo, Kaleem Siddiqi, Dirk B. Walther, Sven Dickinson

Figure-ground organization in the human visual system relies on several shape-based cues, including surroundedness, convexity, and symmetry. While these cues have been extensively studied using abstract stimuli, little is known about how they operate under natural conditions or h…

View free PDFSource page
arxivcs.CV2026-06-26

Text as Illumination: Spatial Contrastive Retinex Learning for Language-guided Medical Image Segmentation

Jian Shi, Cheng Zhen, Pingping Zhang, Rui Xu, Yanan Lv, Yili Ma, et al.

Language-guided Medical Image Segmentation (LMIS) has shown great potential to improve the delineation of anatomical structures and lesions by integrating clinical textual information. Existing methods generally rely on either implicit interaction between textual and visual featu…

View free PDFSource page
arxivcs.CV2026-07-15

MixCompress: Mixture of Experts for Variable Rate Learned Image Compression

Calvin-Khang Ta, Praneet Singh, Tong Shao, Peng Yin

Learned image compression (LIC) is bottlenecked by the need to store independent models for each rate-distortion operating point. Existing variable bit-rate (VBR) methods aim to reduce this overhead via dense parameter modulation, but forcing a shared backbone to approximate dive…

View free PDFSource page
arxivcs.CV2026-07-10

Probing Diffusion Denoising Dynamics for Contrastive Representation Learning

Yasong Dai, Zeeshan Hayder, David Ahmedt-Aristizabal, Hongdong Li

Text-to-image diffusion models exhibit unprecedented generative capability and contain rich intermediate representations that can be useful for discriminative vision tasks. Motivated by this observation, we study a focused question: how can the denoising dynamics of a pretrained…

View free PDFSource page
arxiveess.IVcs.CV2026-07-21

Wavefront Parallelization for Efficient Learned Image Compression

Shimon Murai, Fangzheng Lin, Kasidis Arunruangsirilert, Jiro Katto

Autoregressive context models are foundational for learned image compression,but they suffer from slow serial inference. Existing acceleration methods such as checkerboard context require architectural changes and retraining, thus are inapplicable to pre-trained models. We propos…

View free PDFSource page