AI Paper Atlas

Research map / Benchmarks / CIFAR-100 (fine-tuning)

CIFAR-100 (fine-tuning) leaderboard

11 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on CIFAR-100 (fine-tuning). Best accuracy: 91.9 (SBM-B, CVPR 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.

CIFAR-100 (fine-tuning) · accuracy

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

#accuracy ↑MethodPaperTopic
191.9SBM-B
ImageNet-1K pretrained
Star with Bilinear Mapping
CVPR 2025 · Table 2
Attention and transformers
291.13HQ+LSS (INT4)
ViT-L/32
Training Transformers with 4-bit Integers
NeurIPS 2023 · Table 1
Quantization
390.4Ours (8 layers)
DeiT-B
Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective
ICML 2026 · Table 3
Attention and transformers
490.1iBOT+AttMask
ViT-S/16
What to Hide from Your Students: Attention-Guided Masked Image Modeling
ECCV 2022 · Tables 6-8
Attention and transformers
589.65LE-SAM
ViT-B-16 (ImageNet-1K pretrained)
Fix the Loss, Not the Radius: Rethinking the Adversarial Perturbation of Sharpness-Aware Minimization
ICML 2026 · Table 3
Implicit bias and generalization
689.58RSAM (Renyi sharpness)
ViT-B-16 (ImageNet-1K pretrained)
Rényi Sharpness: A Novel Sharpness that Strongly Correlates with Generalization
ICLR 2026 · Table 2
Implicit bias and generalization
789DiSKDiSK: Differentially Private Optimizer with Simplified Kalman Filter for Noise Reduction
ICLR 2025 · Contributions
Differential privacy
884.82SmartFRZ
DeiT-T
SmartFRZ: An Efficient Training Framework using Attention-Based Layer Freezing
ICLR 2023 · Table 3
RNNs and normalization
984.21p-frac (1 image, LPCE)
ViT-Tiny
Scaling Backwards: Minimal Synthetic Pre-training?
ECCV 2024 · Table 2
Pruning and efficient tuning
1080.3CRATE-B
ImageNet-1K pre-trained
Masked Completion via Structured Diffusion with White-Box Transformers
ICLR 2024 · Classification accuracy table
RNNs and normalization
1 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.