SUN397

Overview edit

Variants: SUN397

Associated Benchmarks

This dataset is used in 6 benchmarks:

• Few-Shot Learning - Metrics: Harmonic mean
• Zero-Shot Learning - Metrics: Accuracy
• Image Clustering - Metrics: Accuracy
• Image Classification - Metrics: Accuracy
• Fine-Grained Image Classification - Metrics: Accuracy
• Prompt Engineering - Metrics: Harmonic mean

Recent Benchmark Submissions

Research Papers

Recent papers with results on this dataset:

• MMRL++: Parameter-Efficient and Interaction-Aware Representation Learning for Vision-Language Models (2025) -
• MMRL: Multi-Modal Representation Learning for Vision-Language Models (2025) -
• HPT++: Hierarchically Prompting Vision-Language Models with Multi-Granularity Knowledge Generation and Improved Structure Modeling (2024) -
• Let Go of Your Labels with Unsupervised Transfer (2024) -
• Label Propagation for Zero-shot Classification with Vision-Language Models (2024) -
• Prompt Learning via Meta-Regularization (2024) -
• PromptKD: Unsupervised Prompt Distillation for Vision-Language Models (2024) -
• Learning Hierarchical Prompt with Structured Linguistic Knowledge for Vision-Language Models (2023) -
• DePT: Decoupled Prompt Tuning (2023) -
• Read-only Prompt Optimization for Vision-Language Few-shot Learning (2023) -
• Self-regulating Prompts: Foundational Model Adaptation without Forgetting (2023) -
• Consistency-guided Prompt Learning for Vision-Language Models (2023) -
• Learning Domain Invariant Prompt for Vision-Language Models (2022) -
• MaPLe: Multi-modal Prompt Learning (2022) -
• Bayesian Prompt Learning for Image-Language Model Generalization (2022) -
• TransBoost: Improving the Best ImageNet Performance using Deep Transduction (2022) -
• An Evolutionary Approach to Dynamic Introduction of Tasks in Large-scale Multitask Learning Systems (2022) -
• Bamboo: Building Mega-Scale Vision Dataset Continually with Human-Machine Synergy (2022) -
• Conditional Prompt Learning for Vision-Language Models (2022) -
• Vision Models Are More Robust And Fair When Pretrained On Uncurated Images Without Supervision (2022) -