RESEARCH PAPERYear 2026

DINOv3

View all authors

Oriane Siméoni; Huy V. Vo; Maximilian Seitzer; Federico Baldassarre; Maxime Oquab; Cijo Jose; Vasil Khalidov; Marc Szafraniec; Seungeun Yi; Michaël Ramamonjisoa; Francisco Massa; Daniel Haziza; Luca Wehrstedt; Jianyuan Wang; Timothée Darcet; Théo Moutakanni; Leonel Sentana; Claire Roberts; Andrea Vedaldi; Jamie Tolan; John Brandt; Camille Couprie; Julien Mairal; Hervé Jégou; Patrick Labatut; Piotr Bojanowski

Read paper PDF Code

Classification

View four quadrants

Major category: Components of WAMs
Quadrant: Not applicable
Architecture: Not applicable
Prediction paradigm: Not applicable
Subcategories: Visual encoders & representations
Source review status: Verified from primary sources

Category review. DINOv3 is a general self-supervised image encoder producing reusable global and dense patch features. Gram anchoring improves representation training and is absent at inference; separate heads consume the frozen features. This matches the visual-encoder component boundary. Reading evidence

AT A GLANCE

Contribution

A contribution summary has not been added yet.

Abstract

An abstract has not been added yet.

Affiliations

Not listed in the collection.

BibTeX

@article{dinov3,
title={{DINO}v3},
author={Oriane Siméoni and Huy V. Vo and Maximilian Seitzer and Federico Baldassarre and Maxime Oquab and Cijo Jose and Vasil Khalidov and Marc Szafraniec and Seungeun Yi and Michaël Ramamonjisoa and Francisco Massa and Daniel Haziza and Luca Wehrstedt and Jianyuan Wang and Timothée Darcet and Théo Moutakanni and Leonel Sentana and Claire Roberts and Andrea Vedaldi and Jamie Tolan and John Brandt and Camille Couprie and Julien Mairal and Hervé Jégou and Patrick Labatut and Piotr Bojanowski},
journal={TMLR},
issn={2835-8856},
year={2026},
}

Classification

Contribution

Abstract

Affiliations

BibTeX

More in this category

V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learning

World Simulation with Video Foundation Models for Physical AI

Seedance 1.0: Exploring the Boundaries of Video Generation Models