mlsanityData to 7 Oct 2026

Attributed Synthetic Data Generation for Zero-shot Domain-specific Image Classification

arXiv 2504.04510 ↗First posted 6 Apr 2025cs.CVcited by 0

On the map: Synthetic data and self-improvement for LLMs → Self-improving MLLMs and synthetic data

Abstract (arXiv)

Zero-shot domain-specific image classification is challenging in classifying real images without ground-truth in-domain training examples. Recent research involved knowledge from texts with a text-to-image model to generate in-domain training images in zero-shot scenarios. However, existing methods heavily rely on simple prompt strategies, limiting the diversity of synthetic training images, thus leading to inferior performance compared to real images. In this paper, we propose AttrSyn, which leverages large language models to generate attributed prompts. These prompts allow for the generation of more diverse attributed synthetic images. Experiments for zero-shot domain-specific image classification on two fine-grained datasets show that training with synthetic images generated by AttrSyn significantly outperforms CLIP's zero-shot classification under most situations and consistently surpasses simple prompt strategies.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Self-improving MLLMs and synthetic data, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 29 of its 39 references that are in the map. First posted on arXiv 6 Apr 2025.

Builds on

29 of 39 references.

References · 29

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  5. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  6. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  7. 2023GPT-4 Technical Reportcited by 18,132
  8. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  9. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  10. 2023SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesiscited by 4,290
  11. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  12. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
17 more
  1. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  2. 2022Diffusion Models: A Comprehensive Survey of Methods and Applicationscited by 1,107
  3. 2022Visual Classification via Description from Large Language Modelscited by 331
  4. 2022Is synthetic data from generative models ready for image recognition?cited by 320
  5. 2022What does a platypus look like? Generating customized prompts for zero-shot image classificationcited by 263
  6. 2023Large Language Model as Attributed Training Data Generator: A Tale of Diversity and Biascited by 230
  7. 2017Augmented Reality Meets Computer Vision : Efficient Data Generation for Urban Driving Scenescited by 122
  8. 2018Learning Semantic Segmentation from Synthetic Data: A Geometrically Guided Input-Output Adaptation Approachcited by 119
  9. 2023Scaling Laws of Synthetic Images for Model Training ... for Nowcited by 107
  10. 2022NeRF-Supervision: Learning Dense Object Descriptors from Neural Radiance Fieldscited by 86
  11. 2023Diversity is Definitely Needed: Improving Model-Agnostic Zero-shot Classification via Stable Diffusioncited by 81
  12. 2015Zero-Shot Learning via Semantic Similarity Embeddingcited by 74
  13. 2024Comprehensive Exploration of Synthetic Data Generation: A Surveycited by 59
  14. 2023Diversify, Don't Fine-Tune: Scaling Up Visual Recognition Training with Synthetic Imagescited by 20
  15. 2023Synthetic Sample Selection for Generalized Zero-Shot Learningcited by 14
  16. 2024Is Synthetic Image Useful for Transfer Learning? An Investigation into Data Generation, Volume, and Utilizationcited by 6
  17. 2017On the Diversity of Realistic Image Synthesiscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Will It Zero-Shot?: Predicting Zero-Shot Classification Performance For Arbitrary Queries
  2. 2024Discriminative Image Generation with Diffusion Models for Zero-Shot Learning
  3. 2022VL-Taboo: An Analysis of Attribute-based Zero-shot Capabilities of Vision-Language Models
  4. 2026Attributes Should Come from Images, Not Class Names: Distribution-Conditioned Attribute Selection for Vision-Language Models
  5. 2025Training-Free Synthetic Data Generation with Dual IP-Adapter Guidance
  6. 2024Diverse and Tailored Image Generation for Zero-shot Multi-label Classification
  7. 2023Text-to-Image Diffusion Models are Zero-Shot Classifiers
  8. 2024Zero-Shot Distillation for Image Encoders: How to Make Effective Use of Synthetic Data