Data to 14 Oct 2026

Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?

arXiv 2003.11539 ↗First posted 25 Mar 2020cs.CV cs.LGcited by 437

Not on the map. Why?

Cited by 21 in Multimodal LLMs / vision-language models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The focus of recent meta-learning research has been on the development of learning algorithms that can quickly adapt to test time tasks with limited data and low computational cost. Few-shot learning is widely used as one of the standard benchmarks in meta-learning. In this work, we show that a simple baseline: learning a supervised or self-supervised representation on the meta-training set, followed by training a linear classifier on top of this representation, outperforms state-of-the-art few-shot learning methods. An additional boost can be achieved through the use of self-distillation. This demonstrates that using a good learned embedding model can be more effective than sophisticated meta-learning algorithms. We believe that our findings motivate a rethinking of few-shot image classification benchmarks and the associated role of meta-learning algorithms. Code is available at: http://github.com/WangYueFt/rfs/.

Key facts. Cited by 437 papers in the map (data through 14 Oct 2026). First posted on arXiv 25 Mar 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 117; 2022: 122; 2023: 95; 2024: 47; 2025: 36; 2026: 20.

437 papers.

Sort:

By area:

2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
2021Learning to Prompt for Vision-Language Modelscited by 2,281
2021Intriguing Properties of Vision Transformerscited by 446
2021Tip-Adapter: Training-free CLIP-Adapter for Better Vision-Language Modelingcited by 431
2024The Platonic Representation Hypothesiscited by 382
2022Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classificationcited by 362
2022Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classificationcited by 312
2022MultiMAE: Multi-modal Multi-task Masked Autoencoderscited by 291
2022Prompt Distribution Learningcited by 235
2022CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understandingcited by 198
2021FedBABU: Towards Enhanced Representation for Federated Image Classificationcited by 182

38 more

2023StableRep: Synthetic Images from Text-to-Image Models Make Strong Visual Representation Learnerscited by 177
2022A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunitiescited by 163
2022Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Differencecited by 126
20234M: Massively Multimodal Masked Modelingcited by 120
2021Using Shape to Categorize: Low-Shot Learning with an Explicit Shape Biascited by 118
2021Overcoming Catastrophic Forgetting in Incremental Few-Shot Learning by Finding Flat Minimacited by 118
2021Simpler is Better: Few-shot Semantic Segmentation with Classifier Weight Transformercited by 116
2021PartImageNet: A Large, High-Quality Dataset of Partscited by 107
2021DeFRCN: Decoupled Faster R-CNN for Few-Shot Object Detectioncited by 106
2021On the Role of Neural Collapse in Transfer Learningcited by 100
2022Constrained Few-shot Class-incremental Learningcited by 94
2022Learning from Few Examples: A Summary of Approaches to Few-Shot Learningcited by 86
2021Cross-domain Few-shot Learning with Task-specific Adapterscited by 80
2021A Broad Study on the Transferability of Visual Representations with Contrastive Learningcited by 78
2021Elaborative Rehearsal for Zero-shot Action Recognitioncited by 77
2021Relational Embedding for Few-Shot Classificationcited by 74
2021Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learningcited by 72
2022Can Foundation Models Perform Zero-Shot Task Specification For Robot Manipulation?cited by 70
2023BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIPcited by 67
2023Advances and Challenges in Meta-Learning: A Technical Reviewcited by 65
2021A Hierarchical Transformation-Discriminating Generative Model for Few Shot Anomaly Detectioncited by 60
2021Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot Learningcited by 56
2021Universal Representation Learning from Multiple Domains for Few-shot Classificationcited by 56
2021Subspace Regularizers for Few-Shot Class Incremental Learningcited by 56
2022HyperTransformer: Model Generation for Supervised and Semi-Supervised Few-Shot Learningcited by 55
2021Factors of Influence for Transfer Learning across Diverse Appearance Domains and Task Typescited by 54
2021Rectifying the Shortcut Learning of Background for Few-Shot Learningcited by 51
2021Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Datacited by 50
2021Bridging Multi-Task Learning and Meta-Learning: Towards Efficient Training and Effective Adaptationcited by 50
2022Joint Distribution Matters: Deep Brownian Distance Covariance for Few-Shot Classificationcited by 50
2023STUNT: Few-shot Tabular Learning with Self-generated Tasks from Unlabeled Tablescited by 49
2023A Closer Look at Few-shot Classification Againcited by 46
2021Improved Regularization and Robustness for Fine-tuning in Neural Networkscited by 45
2021Orthogonal Projection Losscited by 44
2021Curious Representation Learning for Embodied Intelligencecited by 44
2022Learning to Detect Mobile Objects from LiDAR Scans Without Labelscited by 44
2021Universal-Prototype Enhancing for Few-Shot Object Detectioncited by 43
2021von Mises-Fisher Loss: An Exploration of Embedding Geometries for Supervised Learningcited by 42

50 most cited of 437 All 437 (JSON)