# Provably Improving Generalization of Few-Shot Models with Synthetic Data

arXiv: https://arxiv.org/abs/2505.24190 · HTML: https://mlsanity.com/paper/2505.24190/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 45 references that are in the map.
- First posted on arXiv 30 May 2025.

## Builds on (most cited first, 30 of 30)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features](https://mlsanity.com/paper/1905.04899/) (2019)
- [Learning to Prompt for Vision-Language Models](https://mlsanity.com/paper/2109.01134/) (2021)
- [Fine-Grained Visual Classification of Aircraft](https://mlsanity.com/paper/1306.5151/) (2013)
- [Visual Prompt Tuning](https://mlsanity.com/paper/2203.12119/) (2022)
- [EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification](https://mlsanity.com/paper/1709.00029/) (2017)
- [Dataset Condensation with Distribution Matching](https://mlsanity.com/paper/2110.04181/) (2021)
- [Self-regulating Prompts: Foundational Model Adaptation without Forgetting](https://mlsanity.com/paper/2307.06948/) (2023)
- [Fake it till you make it: Learning transferable representations from synthetic ImageNet clones](https://mlsanity.com/paper/2212.08420/) (2022)
- [Prompt, Generate, then Cache: Cascade of Foundation Models makes Strong Few-shot Learners](https://mlsanity.com/paper/2303.02151/) (2023)
- [Improved Distribution Matching for Dataset Condensation](https://mlsanity.com/paper/2307.09742/) (2023)
- [Robust Classification with Convolutional Prototype Learning](https://mlsanity.com/paper/1805.03438/) (2018)
- [D4M: Dataset Distillation via Disentangled Diffusion Model](https://mlsanity.com/paper/2407.15138/) (2024)
- [TCP:Textual-based Class-aware Prompt tuning for Visual-Language Model](https://mlsanity.com/paper/2311.18231/) (2023)
- [PromptKD: Unsupervised Prompt Distillation for Vision-Language Models](https://mlsanity.com/paper/2403.02781/) (2024)
- [MMRL: Multi-Modal Representation Learning for Vision-Language Models](https://mlsanity.com/paper/2503.08497/) (2025)
- [Real-Fake: Effective Training Data Synthesis Through Distribution Matching](https://mlsanity.com/paper/2310.10402/) (2023)
- [Synthetic data, real errors: how (not) to publish and use synthetic data](https://mlsanity.com/paper/2305.09235/) (2023)
- [DataDream: Few-shot Guided Dataset Generation](https://mlsanity.com/paper/2407.10910/) (2024)
- [Toward Understanding Generative Data Augmentation](https://mlsanity.com/paper/2305.17476/) (2023)
- [High-dimensional Analysis of Knowledge Distillation: Weak-to-Strong Generalization and Scaling Laws](https://mlsanity.com/paper/2410.18837/) (2024)
- [Large Language Models are Good Prompt Learners for Low-Shot Image Classification](https://mlsanity.com/paper/2312.04076/) (2023)
- [Towards a Theoretical Understanding of Synthetic Data in LLM Post-Training: A Reverse-Bottleneck Perspective](https://mlsanity.com/paper/2410.01720/) (2024)
- [A Bag-of-Prototypes Representation for Dataset-Level Applications](https://mlsanity.com/paper/2303.13251/) (2023)
- [Diversified in-domain synthesis with efficient fine-tuning for few-shot classification](https://mlsanity.com/paper/2312.03046/) (2023)
- [Prototype-based Dataset Comparison](https://mlsanity.com/paper/2309.02401/) (2023)
- [Gentle Local Robustness implies Generalization](https://mlsanity.com/paper/2412.06381/) (2024)
- [A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasets](https://mlsanity.com/paper/2402.03985/) (2024)

## Built on by (most cited first, 3 of 3)

- [DTGen: Generative Diffusion-Based Few-Shot Data Augmentation for Fine-Grained Dirty Tableware Recognition](https://mlsanity.com/paper/2509.11661/) (2025)
- [Using Synthetic Data to estimate the True Error is theoretically and practically doable](https://mlsanity.com/paper/2511.00964/) (2025)
- [When Hard Negatives Hurt: Bridging the Generative-Discriminative Gap in Hard Negative Synthesis for Retrieval](https://mlsanity.com/paper/2606.01304/) (2026)

## Abstract (arXiv)

Few-shot image classification remains challenging due to the scarcity of labeled training examples. Augmenting them with synthetic data has emerged as a promising way to alleviate this issue, but models trained on synthetic samples often face performance degradation due to the inherent gap between real and synthetic distributions. To address this limitation, we develop a theoretical framework that quantifies the impact of such distribution discrepancies on supervised learning, specifically in the context of image classification. More importantly, our framework suggests practical ways to generate good synthetic samples and to train a predictor with high generalization ability. Building upon this framework, we propose a novel theoretical-based algorithm that integrates prototype learning to optimize both data partitioning and model training, effectively bridging the gap between real few-shot data and synthetic data. Extensive experiments results show that our approach demonstrates superior performance compared to state-of-the-art methods, outperforming them across multiple datasets.