AI Paper Atlas

Research map / Benchmarks / CIFAR-100 linear eval

CIFAR-100 linear eval leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on CIFAR-100 linear eval. Best accuracy: 72.4 (S-TEC*, NeurIPS 2022).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

CIFAR-100 linear eval · accuracy

10 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
172.4S-TEC*
ResNet-50
Self-Supervised Learning Through Efference Copies
NeurIPS 2022 · Table 1
Self-supervised and contrastive
272.17All4One
ResNet-18
All4One: Symbiotic Neighbour Contrastive Learning via Self-Attention and Redundancy Reduction
ICCV 2023 · Table 1
Self-supervised and contrastive
371.39BYOL + PairDist
ConvNeXt
An Information Theory Perspective on Variance-Invariance-Covariance Regularization
NeurIPS 2023 · Table 1
Self-supervised and contrastive
470.65MIRA
ResNet-18
Unsupervised Visual Representation Learning via Mutual Information Regularized Assignment
NeurIPS 2022 · Table 7
Recognition and long-tail
569.5IE-CL
ResNet-18
Maximizing Incremental Information Entropy for Contrastive Learning
ICLR 2026 · Table 2a
Self-supervised and contrastive
667.53BYOL + GenView
ResNet-18
GenView: Enhancing View Quality with Pretrained Generative Model for Self-Supervised Learning
ECCV 2024 · Table 5
Self-supervised and contrastive
764.38SimSiam + Hallucinator
ResNet-18
Hallucination Improves the Performance of Unsupervised Visual Representation Learning
ICCV 2023 · Table 1
Self-supervised and contrastive
862.72MoCov2 + OA-Crop (CAM)
ResNet-50
Object-aware Contrastive Learning for Debiased Scene Representation
NeurIPS 2021 · Table 3
Open-set and few-shot detection
960.1PMSN power-law prior (tau=0.25) on iNat18
ViT-S/16
The hidden uniform cluster prior in self-supervised learning
ICLR 2023 · Table 3
Self-supervised and contrastive
1049.79TSPipe
BYOL ResNet-18
TSPipe: Learn from Teacher Faster with Pipelines
ICML 2022 · Table 4
Knowledge distillation

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.