AI Paper Atlas

Research map / Benchmarks / Domain generalization, ImageNet-A

Domain generalization, ImageNet-A leaderboard

12 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on Domain generalization, ImageNet-A. Best accuracy: 51.97 (VaMP, NeurIPS 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.

Domain generalization, ImageNet-A · accuracy

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

#accuracy ↑MethodPaperTopic
151.97VaMP
CLIP ViT-B/16
VaMP: Variational Multi-Modal Prompt Learning for Vision-Language Models
NeurIPS 2025 · Domain generalization table
CLIP and zero-shot
251.8DeAR
CLIP ViT-B/16
DeAR: Fine-Grained VLM Adaptation by Decomposing Attention Head Roles
CVPR 2026 · Table 4
CLIP and zero-shot
351.64COMMA
CLIP ViT-B/16
COMMA: Co-articulated Multi-Modal Learning
AAAI 2024 · Table: Comparison with existin
CLIP and zero-shot
451.33SDSA
CLIP ViT-B/16
SDSA: Shallow-Deep Squeezing Adapter for Vision-Language Models
ECCV 2026 · Table 3
Cross-modal learning
551MMRL++
CLIP ViT-B/16
MMRL++: Parameter-Efficient and Interaction-Aware Representation Learning for Vision-Language Models
IJCV 2025 · Table 4
Multimodal LLMs
650.94ALIGN
CLIP ViT-B/16
Tuning Multi-mode Token-level Prompt Alignment across Modalities
NeurIPS 2023 · Table 3
CLIP and zero-shot
750.94Ours (CTP + TFT)
CLIP ViT-B/16
Task-Oriented Multi-Modal Mutual Leaning for Vision-Language Models
ICCV 2023 · Domain generalization table
CLIP and zero-shot
850.9MaPLe
CLIP ViT-B/16
MaPLe: Multi-Modal Prompt Learning
CVPR 2023 · Table: Domain generalization
CLIP and zero-shot
950.67CoOp + CaPTA Closer Look at Conditional Prompt Tuning for Vision-Language Models
IJCV 2026 · Table 7
CLIP and zero-shot
1050.28Promise
CLIP ViT-B/16
Prompt-Robust Vision-Language Models via Meta-Finetuning
ICLR 2026 · Table 6
CLIP and zero-shot
2 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.