AI Paper Atlas

Research map / Benchmarks / ImageNet ReaL

ImageNet ReaL leaderboard

21 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ImageNet ReaL. Best accuracy: 91.1 (MAWS ViT-6.5B, ICCV 2023).

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.

ImageNet ReaL · accuracy

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

#accuracy ↑MethodPaperTopic
191.1MAWS ViT-6.5B
ViT-6.5B, IG-3B
The Effectiveness of MAE Pre-Pretraining for Billion-Scale Pretraining
ICCV 2023 · Table 3
CLIP and zero-shot
290.81ViT-G/14Scaling Vision Transformers
CVPR 2022 · Table: ViT-G/14 results
Attention and transformers
390.72ViT-H/14 (JFT)An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
ICLR 2021 · Table 1
CNN design
490.66ViT-Calibrator (+All) on Eva-L
Eva-L
ViT-Calibrator: Decision Stream Calibration for Vision Transformer
AAAI 2024 · Table 1
Attention and transformers
590.6VOLO-D5 (finetune 512)VOLO: Vision Outlooker for Visual Recognition
TPAMI 2022 · Table 1
Attention and transformers
690.2CaiT-M48↑448 distilledGoing Deeper With Image Transformers
ICCV 2021 · Table (complexity vs accuracy)
Attention and transformers
788.6LipsFormer-CSwin-B
CSwin-B
LipsFormer: Introducing Lipschitz Continuity to Vision Transformers
ICLR 2023 · Appendix: Overfitting Evaluati
Attention and transformers
888.2JFT memory
DinoV2 ViT-L14
Towards flexible perception with visual memory
ICML 2025 · Table: OOD evaluation
Human and embodied modeling
988ARM-B
MambaMLP-B
Autoregressive Pretraining with Mamba in Vision
ICLR 2025 · Robustness and Generalization
Attention and transformers
1087.3CeiT-SIncorporating Convolution Designs Into Visual Transformers
ICCV 2021 · Table 4
Attention and transformers
11 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

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