AI Paper Atlas

Research map / Benchmarks / GTSRB

GTSRB leaderboard

15 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on GTSRB. Best accuracy: 99.2 (Random Masking (1%), ICML 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.

GTSRB · accuracy

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

#accuracy ↑MethodPaperTopic
199.2Random Masking (1%)
ViT-B/16
Random Masking Finds Winning Tickets for Parameter Efficient Fine-tuning
ICML 2024 · Image classification table
Pruning and efficient tuning
298.73Auto-Switch
CLIP ViT-B/32
Less is More: Efficient Model Merging with Binary Task Switch
CVPR 2025 · Table 2
Pruning and efficient tuning
398.1VGP, all prompts
ViG-M
Vision Graph Prompting via Semantic Low-Rank Decomposition
ICML 2025 · Table 3
GNN architectures
497.96GeN-SGD
ResNet50
Gradient descent with generalized Newton’s method
ICLR 2025 · Image classification table
Optimizers
595.9Full context-agnosticTowards Context-Agnostic Learning Using Synthetic Data
NeurIPS 2021 · Table: GTSRB results
Recognition and long-tail
695.66LoRA-Pro
CLIP-ViT-B/16
LoRA-Pro: Are Low-Rank Adapters Properly Optimized?
ICLR 2025 · Image classification results t
Pruning and efficient tuning
794.9CLIP ViT-L/14@336px
ViT-L/14
Learning Transferable Visual Models From Natural Language Supervision
ICML 2021 · Appendix linear-probe results
CLIP and zero-shot
894.25SPU
CLIP-ViT/B-16
Overcoming Generic Knowledge Loss with Selective Parameter Update
CVPR 2024 · Table 1
Continual fine-tuning of LLMs
994.1VPT+TCPA
ViT-B/16 (ImageNet-21k)
Token Coordinated Prompt Attention is Needed for Visual Prompting
ICML 2025 · Table 3
CLIP and zero-shot
1092.01Finetune with Contrastive Regularization
CLIP ViT-L
The Trade-off between Universality and Label Efficiency of Representations from Contrastive Learning
ICLR 2023 · Foundation-model results table
Self-supervised and contrastive
5 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.