AI Paper Atlas

Research map / Benchmarks / ImageNette

ImageNette leaderboard

17 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on ImageNette. Best accuracy: 99.59 (AutoGnothi surrogate, ICLR 2025).

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.

ImageNette · accuracy

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

#accuracy ↑MethodPaperTopic
199.59AutoGnothi surrogate
ViT-base
Gnothi Seauton: Empowering Faithful Self-Interpretability in Black-Box Transformers
ICLR 2025 · Training and memory efficiency
Recognition and long-tail
299.3DistDiff expansion 5×
ImageNet-1k pretrained ResNet-50
Distribution-Aware Data Expansion with Diffusion Models
NeurIPS 2024 · Table: Comparison of using str
Fast diffusion sampling
398.99SKITTesting Semantic Importance via Betting
NeurIPS 2024 · Table 1
Recognition and long-tail
498.7BC-LLM
GPT-4o-mini
Bayesian Concept Bottleneck Models with LLM Priors
NeurIPS 2025 · Appendix, Other image datasets
Language, safety and interpretability
593.2GEQ (13M)GEQ: Gaussian Kernel Inspired Equilibrium Models
NeurIPS 2023 · Table 2
Kernels and regression theory
689.3HoResNet-18
ResNet-18
Convolution Goes Higher-Order: A Biologically Inspired Mechanism Empowers Image Classification
NeurIPS 2025 · Imagenette classification tabl
CNN design
789.04RT defense, no attack
ResNet-34
Demystifying the Adversarial Robustness of Random Transformation Defenses
ICML 2022 · Table 2
Certified and empirical robustness
888.7VIKING
ResNet-34
VIKING: Deep variational inference with stochastic projections
NeurIPS 2025 · Table 1
Uncertainty and Gaussian processes
986.8LICO
Wide ResNet-18
Reproducibility study of “LICO: Explainable Models with Language-Image Consistency"
NeurIPS 2024 · Test accuracy table
CLIP and zero-shot
1081.1Ours (DATM)Post Training Quantization for Efficient Dataset Condensation
AAAI 2026 · Table 3
Quantization
7 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.