AI Paper Atlas

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

All4One: Symbiotic Neighbour Contrastive Learning via Self-Attention and Redundancy Reduction

Imanol G. Estepa, Ignacio Sarasua, Bhalaji Nagarajan, Petia Radeva

ICCV 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Contrastive versus non-contrastive theory · 13 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
add centroid objective to NNCLR objective (ablation)All4One · CIFAR-100top-1 accuracy: 69.62 -> 71.02Improved
add EMA to NNCLR + centroid (ablation)All4One · CIFAR-100top-1 accuracy: 71.02 -> 71.08No clear effect
add feature contrast (redundancy reduction) to NNCLR + centroid + EMA (ablation)All4One · CIFAR-100top-1 accuracy: 71.08 -> 72.17Improved
add neighbour objective to centroid + feature + EMA (ablation)All4One · CIFAR-100top-1 accuracy: 71.31 -> 72.17Improved
add centroid objective to NNCLR + feature + EMA (ablation)All4One · CIFAR-100top-1 accuracy: 71.64 -> 72.17Improved
combine feature contrast with neighbour contrast to cut embedding dimension from 2048 to 256 (ablation)All4One vs Barlow Twins · CIFAR-100top-1 accuracy: Barlow Twins 71.21 (2048) -> 62.14 (256); All4One (256) 72.17Improved
use All4One under crop-only augmentation instead of NNCLR (ablation)All4One · CIFAR-100 (weak augmentation)top-1 accuracy: 35.39 -> 44.7Improved
increase transformer encoder layers from 3 to 9 (ablation)All4One · CIFAR-100top-1 accuracy: 72.17 -> 71.75Hurt
increase number of neighbours from 5 to 20 (ablation)All4One · CIFAR-100top-1 accuracy: 72.17 -> 71.79Hurt
replace NNCLR with All4One on Pets transferAll4One · Pets transfertop-1 accuracy: 83.34 -> 81.57Hurt
higher NN retrieval accuracy does not imply better accuracy (ablation)All4One · CIFAR-100NN retrieval vs top-1: 86.16 (v3) vs 82.16 (v6, best top-1 72.17)No clear effect

Later papers that compared with this method

1 later accepted paper reports All4One as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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