AI Paper Atlas

Research map / Benchmarks / ImageNet-100

ImageNet-100 leaderboard

73 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ImageNet-100. Best accuracy: 92.5 (ConvNeXt-T+DiTAC, ECCV 2024).

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-100 · accuracy

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

#accuracy ↑MethodPaperTopic
192.5ConvNeXt-T+DiTAC
ConvNeXt-T
Trainable Highly-expressive Activation Functions
ECCV 2024 · Table 4
RNNs and normalization
291.92FoCA CBM with iterative refinement
ResNet
Formal Concept Lattices are Good Semantic Scaffolds for Concept-Based Learning
ICML 2026 · Table A5
Recognition and long-tail
391.44VIPER
ViT-B/16
Exposing Functional Fusion: A New Class of Strategic Backdoor in Dynamic Prompt Architectures
CVPR 2026 · Table 3
Backdoor and poisoning
490.9+SoftHGNN-SeS
SwinT-Tiny
SoftHGNN: Soft Hypergraph Neural Networks for General Visual Recognition
IJCV 2026 · Table 1
Recognition and long-tail
590.7ConvNeXt-t+AIM[2, 35%]
ConvNeXt-tiny
AIM: Amending Inherent Interpretability via Self-Supervised Masking
ICCV 2025 · Table 1
Self-supervised and contrastive
690.3LTG (Epistemic)
ResNet-34
Generative Data Mining with Longtail-Guided Diffusion
ICML 2025 · ImageNet variants table
Autoregressive and latent generation
790.3VisualAtom-1k
ViT-T
Visual Atoms: Pre-Training Vision Transformers With Sinusoidal Waves
CVPR 2023 · Table: comparison of pre-train
Pruning and efficient tuning
889.32A-VARC+
VAR-d16
Your VAR Model is Secretly an Efficient and Explainable Generative Classifier
ICLR 2026 · Table 1
Fast diffusion sampling
989.2ExFractalDB-1k*Replacing Labeled Real-Image Datasets With Auto-Generated Contours
CVPR 2022 · Table 9
Self-supervised and contrastive
1088.7SelfCon
ResNet-50
Self-Contrastive Learning: Single-Viewed Supervised Contrastive Framework Using Sub-network
AAAI 2023 · Table 2
Self-supervised and contrastive
63 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.