mlsanityData to 7 Oct 2026

FreqGRL: Suppressing Low-Frequency Bias and Mining High-Frequency Knowledge for Cross-Domain Few-Shot Learning

arXiv 2511.06648 ↗First posted 10 Nov 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Cross-domain few-shot learning (CD-FSL) aims to recognize novel classes with only a few labeled examples under significant domain shifts. While recent approaches leverage a limited amount of labeled target-domain data to improve performance, the severe imbalance between abundant source data and scarce target data remains a critical challenge for effective representation learning. We present the first frequency-space perspective to analyze this issue and identify two key challenges: (1) models are easily biased toward source-specific knowledge encoded in the low-frequency components of source data, and (2) the sparsity of target data hinders the learning of high-frequency, domain-generalizable features. To address these challenges, we propose FreqGRL, a novel CD-FSL framework that mitigates the impact of data imbalance in the frequency space. Specifically, we introduce a Low-Frequency Replacement (LFR) module that substitutes the low-frequency components of source tasks with those from the target domain to create new source tasks that better align with target characteristics, thus reducing source-specific biases and promoting generalizable representation learning. We further design a High-Frequency Enhancement (HFE) module that filters out low-frequency components and performs learning directly on high-frequency features in the frequency space to improve cross-domain generalization. Additionally, a Global Frequency Filter (GFF) is incorporated to suppress noisy or irrelevant frequencies and emphasize informative ones, mitigating overfitting risks under limited target supervision. Extensive experiments on five standard CD-FSL benchmarks demonstrate that our frequency-guided framework achieves state-of-the-art performance.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 33 of its 53 references that are in the map. First posted on arXiv 10 Nov 2025.

Builds on

33 of 53 references.

References · 33

  1. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  2. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  3. 2016Matching Networks for One Shot Learningcited by 2,246
  4. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  5. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
  6. 2020Meta-Learning in Neural Networks: A Surveycited by 944
  7. 2019Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examplescited by 312
  8. 2021Generalizing Face Forgery Detection with High-frequency Featurescited by 261
  9. 2020Learning in the Frequency Domaincited by 196
  10. 2020Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformationcited by 183
  11. 2021FSDR: Frequency Space Domain Randomization for Domain Generalizationcited by 163
  12. 2020Self-training for Few-shot Transfer Across Extreme Task Differencescited by 75
21 more
  1. 2022Improving Vision Transformers by Revisiting High-frequency Componentscited by 75
  2. 2022Masked Frequency Modeling for Self-Supervised Visual Pre-Trainingcited by 70
  3. 2021Cross-Domain Few-Shot Classification via Adversarial Task Augmentationcited by 63
  4. 2020Explanation-Guided Training for Cross-Domain Few-Shot Classificationcited by 58
  5. 2023StyleAdv: Meta Style Adversarial Training for Cross-Domain Few-Shot Learningcited by 53
  6. 2021Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencodercited by 51
  7. 2021Meta-FDMixup: Cross-Domain Few-Shot Learning Guided by Labeled Target Datacited by 48
  8. 2022Adversarial Feature Augmentation for Cross-domain Few-shot Classificationcited by 36
  9. 2023ALOFT: A Lightweight MLP-like Architecture with Dynamic Low-frequency Transform for Domain Generalizationcited by 28
  10. 2021Ranking Distance Calibration for Cross-Domain Few-Shot Learningcited by 27
  11. 2024Flatten Long-Range Loss Landscapes for Cross-Domain Few-Shot Learningcited by 26
  12. 2022Wave-SAN: Wavelet based Style Augmentation Network for Cross-Domain Few-Shot Learningcited by 23
  13. 2022TGDM: Target Guided Dynamic Mixup for Cross-Domain Few-Shot Learningcited by 16
  14. 2021On the Importance of Distractors for Few-Shot Classificationcited by 14
  15. 2019Band-limited Training and Inference for Convolutional Neural Networkscited by 13
  16. 2021Few-Shot Learning by Integrating Spatial and Frequency Representationcited by 11
  17. 2021Cross-domain few-shot learning with unlabelled datacited by 10
  18. 2022ME-D2N: Multi-Expert Domain Decompositional Network for Cross-Domain Few-Shot Learningcited by 9
  19. 2022ReFine: Re-randomization before Fine-tuning for Cross-domain Few-shot Learningcited by 6
  20. 2021Self-Taught Cross-Domain Few-Shot Learning with Weakly Supervised Object Localization and Task-Decompositioncited by 4
  21. 2022Few-shot Fine-grained Image Classification via Multi-Frequency Neighborhood and Double-cross Modulationcited by 4

Built on by

Not cited yet.