AI Paper Atlas

Research map / Benchmarks / ImageNet-1K zero-shot

ImageNet-1K zero-shot leaderboard

28 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ImageNet-1K zero-shot. Best accuracy: 84.1 (FG-CLIP 2, ICML 2026).

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-1K zero-shot · accuracy

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

#accuracy ↑MethodPaperTopic
184.1FG-CLIP 2
ViT-So/16
FG-CLIP 2: A Bilingual Fine-grained Vision-Language Alignment Model
ICML 2026 · Table 4
Human and embodied modeling
283.9IMP-MoE-LAlternating Gradient Descent and Mixture-of-Experts for Integrated Multimodal Perception
NeurIPS 2023 · Table 1
Cross-modal learning
383.8Soft MoE-L/16
ViT-L/16
From Sparse to Soft Mixtures of Experts
ICLR 2024 · Table: LIT-style evaluation
Mixture of experts
483CLIPA-G/14 (I36,T8), DataComp-1B, 512M@224 + 128M@336An Inverse Scaling Law for CLIP Training
NeurIPS 2023 · Table 2
CLIP and zero-shot
582.1MAWS ViT-2B
ViT-2B, IG-3B
The Effectiveness of MAE Pre-Pretraining for Billion-Scale Pretraining
ICCV 2023 · Table 6
CLIP and zero-shot
678.5REACT (ViT-L/14, Web-800M, 6M retrieved pairs)
CLIP ViT-L/14
Learning Customized Visual Models With Retrieval-Augmented Knowledge
CVPR 2023 · Table: zero-shot task transfer
CLIP and zero-shot
777.7OVT-MetaCLIP
MetaCLIP ViT-L/14
Omniview-Tuning: Boosting Viewpoint Invariance of Vision-Language Pre-training Models
ECCV 2024 · Table 2
CLIP and zero-shot
877.21SALM
CLIP ViT-L-14-336
Unlocking Fine-Grained Perception in CLIP via Structurally-Aware Latent Masked Modeling
NeurIPS 2026 · Table 1
CLIP and zero-shot
977.2ALOE
SigLIP2 ViT-B/16
Align Once to Explain: Feature Alignment for Scalable B-cosification of Foundational Vision Transformers
CVPR 2026 · Table 3b
Attention and transformers
1077.1D-iGPT
ViT-L/14
Rejuvenating image-GPT as Strong Visual Representation Learners
ICML 2024 · Table 4
Human and embodied modeling
18 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.