AI Paper Atlas

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

Neural Eigenfunctions are Structured Representation Learners

Zhijie Deng, Jiaxin Shi, Hao Zhang, Peng Cui, Cewu Lu, Jun Zhu

TPAMI 2025 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 9 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
use Neural Eigenmap instead of Barlow TwinsSSL pretraining · ImageNet linear probe, 100 epochstop-1 accuracy: 66.2 -> 68.4Improved
increase batch size from 256 to 2048 (ablation)Neural Eigenmap · ImageNet linear probetop-1 accuracy: 60.5 -> 66.8Improved
increase batch size from 1024 to 2048 (ablation)Barlow Twins · ImageNet linear probetop-1 accuracy: 61.8 -> 60.4Hurt
increase projector dimension from 2048 to 8192 (ablation)Neural Eigenmap · ImageNet linear probetop-1 accuracy: 66.8 -> 68.4Improved
increase projector dimension to 8192 (ablation)SCL · ImageNet linear probetop-1 accuracy: 66.5 -> NaNHurt
add stop-gradient ordered eigenfunction learning vs SCL+PCA for short codesNeural Eigenmap · NUS-WIDE retrieval, code length 4mAP: 0.4845 -> 0.6706Improved
use Neural Eigenmap instead of SCL+PCA for long codesNeural Eigenmap · COCO retrieval, code length 32mAP: 0.6902 -> 0.6832Hurt
train longer from 100 to 400 epochs (ablation)Neural Eigenmap · ImageNet linear probetop-1 accuracy: 68.4 -> 71.5Improved
use Neural Eigenmap instead of Barlow Twins for transferMask R-CNN C4 · COCO detectionAP: 39.2 -> 39.9Improved
use Neural Eigenmap instead of GCN with 10% labelsnode classification · OGBN-Productstest accuracy: 73.14 -> 75.78Improved
use Neural Eigenmap instead of Laplacian Eigenmap + MLPnode classification · OGBN-Products 100% labelstest accuracy: 64.21 -> 78.33Improved
remove MLP projector or BN in hidden layer (ablation)Neural Eigenmap · ImageNetconvergenceImproved

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