AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning

Shashanka Venkataramanan, Valentinos Pariza, Mohammadreza Salehi, Lukas Knobel, Elias Ramzi, Spyros Gidaris, Andrei Bursuc, Yuki M Asano

CVPR 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 27 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Franca recipe (Matryoshka heads, CyclicMask, RASA) vs DINOv2 on same IN-21K data, ViT-L without HRFTDINOv2 · ADE20K in-contextmIoU: 33.5 -> 34.5Improved
Franca vs DINOv2 on DAVIS VOS, ViT-L without HRFT/RASADINOv2 · DAVIS VOSJ&F: 64.2 -> 68.0Improved
Franca vs DINOv2 ViT-B kNN with HRFT+RASADINOv2 · IN-1K kNNaccuracy: 76.8 -> 79.6Improved
Franca (undistilled) vs official distilled DINOv2 ViT-B linear segmentation on VOCDINOv2 (distilled) · VOC linear segmentationmIoU: 90.2 -> 89.4Hurt
distill from ViT-G teacher (official DINOv2 vs IN-21K reproduction) ViT-BDINOv2 · VOC linear segmentationmIoU: 86.9 -> 90.2Improved
Franca ViT-G vs DINOv2-G in-context segmentationDINOv2-G · Pascal VOC in-contextmIoU: 73.7 -> 76.7Improved
add RASA post-training to reduce positional bias (ablation)Franca · patch cluster spatial entropymean spatial entropyImproved
add Matryoshka representations, high-res finetune, RASA incrementally (Fig. 2) (ablation)DINOv2-B IN-21K · IN-1K linear / VOC in-contextlinear acc: 81.2 -> 82.6Improved
Franca vs DINOv2 on Feat2GS geometry probingDINOv2 · Feat2GS GeometryPSNR: 19.34 -> 19.58Improved
Franca vs SigLIP 2 on texture probingSigLIP 2 · Feat2GS TexturePSNR: 18.11 vs 17.97No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy