mlsanityData to 7 Oct 2026

Prompt as Free Lunch: Enhancing Diversity in Source-Free Cross-domain Few-shot Learning through Semantic-Guided Prompting

arXiv 2412.00767 ↗First posted 1 Dec 2024cs.CV cs.CL cs.LGcited by 10

Not on the map. Why?

Abstract (arXiv)

The source-free cross-domain few-shot learning (CD-FSL) task aims to transfer pretrained models to target domains utilizing minimal samples, eliminating the need for source domain data. Addressing this issue requires models to have robust generalization abilities and strong feature representation, aligning with the characteristics of large-scale pretrained models. However, large-scale models tend to lose representational ability in cross-domain scenarios due to limited sample diversity. zlhGiven the abundant diversity provided by semantic modality, this paper leverages textual modality to enhance training sample diversity with CLP model, meanwhile improving model transfer efficiency. Specifically, we propose the SeGD-VPT framework, which is divided into two phases. The first step aims to increase feature diversity by adding diversity prompts to each support sample, thereby generating varying input and enhancing sample diversity. Furthermore, we use diversity descriptions of classes to guide semantically meaningful learning of diversity prompts, proposing random combinations and selections of texts to increase textual diversity. Additionally, deep prompt tuning is introduced to enhance the model's transfer capability. After training of the first step, support samples with different diversity prompts are input into the CLIP backbone to generate enhanced features. After generation, the second phase trains classifiers using the generated features. Extensive experimental results across several benchmarks verify our method is comparable to SOTA source-utilized models and attain the best performance under the source-free CD-FSL setting.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 28 of its 44 references that are in the map. First posted on arXiv 1 Dec 2024.

Builds on

28 of 44 references.

References · 28

  1. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  2. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  3. 2016Matching Networks for One Shot Learningcited by 2,246
  4. 2017EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classificationcited by 1,488
  5. 2021CLIP-Adapter: Better Vision-Language Models with Feature Adapterscited by 1,021
  6. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 969
  7. 2021P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Taskscited by 686
  8. 2019Skin Lesion Analysis Toward Melanoma Detection 2018: A Challenge Hosted by the International Skin Imaging Collaboration (ISIC)cited by 657
  9. 2021Tip-Adapter: Training-free CLIP-Adapter for Better Vision-Language Modelingcited by 429
  10. 2016Using Deep Learning for Image-Based Plant Disease Detectioncited by 216
  11. 2020Cross-Domain Few-Shot Classification via Learned Feature-Wise Transformationcited by 183
  12. 2023Large-scale Multi-Modal Pre-trained Models: A Comprehensive Surveycited by 153
16 more
  1. 2023Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learnerscited by 152
  2. 2022Finetune like you pretrain: Improved finetuning of zero-shot vision modelscited by 149
  3. 2021Targeted Supervised Contrastive Learning for Long-Tailed Recognitioncited by 132
  4. 2022Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Differencecited by 123
  5. 2023LoCoOp: Few-Shot Out-of-Distribution Detection via Prompt Learningcited by 107
  6. 2020Self-training for Few-shot Transfer Across Extreme Task Differencescited by 75
  7. 2021Cross-Domain Few-Shot Classification via Adversarial Task Augmentationcited by 63
  8. 2023PromptStyler: Prompt-driven Style Generation for Source-free Domain Generalizationcited by 59
  9. 2020Explanation-Guided Training for Cross-Domain Few-Shot Classificationcited by 58
  10. 2023Read-only Prompt Optimization for Vision-Language Few-shot Learningcited by 58
  11. 2021Boosting the Generalization Capability in Cross-Domain Few-shot Learning via Noise-enhanced Supervised Autoencodercited by 51
  12. 2021Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Datacited by 49
  13. 2021Meta-FDMixup: Cross-Domain Few-Shot Learning Guided by Labeled Target Datacited by 48
  14. 2023StyLIP: Multi-Scale Style-Conditioned Prompt Learning for CLIP-based Domain Generalizationcited by 36
  15. 2020Feature Transformation Ensemble Model with Batch Spectral Regularization for Cross-Domain Few-Shot Classificationcited by 20
  16. 2022Few-Shot Learning of Compact Models via Task-Specific Meta Distillationcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 5; 2026: 5.

10 papers.

  1. 2025EgoCross: Benchmarking Multimodal Large Language Models for Cross-Domain Egocentric Video Question Answeringcited by 19
  2. 2025EgoNight: Towards Egocentric Vision Understanding at Night with a Challenging Benchmarkcited by 16
  3. 2025Domain-RAG: Retrieval-Guided Compositional Image Generation for Cross-Domain Few-Shot Object Detectioncited by 14
  4. 2026Reclaiming Lost Text Layers for Source-Free Cross-Domain Few-Shot Learningcited by 5
  5. 2026Mind the Discriminability Trap in Source-Free Cross-domain Few-shot Learningcited by 4
  6. 2026Interpretable Cross-Domain Few-Shot Learning with Rectified Target-Domain Local Alignmentcited by 4
  7. 2025VIPAMIN: Visual Prompt Initialization via Embedding Selection and Subspace Expansioncited by 1
  8. 2026Reviving In-domain Fine-tuning Methods for Source-Free Cross-domain Few-shot Learningcited by 1
  9. 2025Multiple Stochastic Prompt Tuning for Few-shot Adaptation under Extreme Domain Shiftcited by 0
  10. 2026When Semantics Saturate or Emerge: Adaptation-Conditional Semantic Utility in Source-Free Cross-Domain Few-Shot Learningcited by 0