# Can Synthetic Images Serve as Effective and Efficient Class Prototypes?

arXiv: https://arxiv.org/abs/2512.17160 · HTML: https://mlsanity.com/paper/2512.17160/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 21 of its 30 references that are in the map.
- First posted on arXiv 19 Dec 2025.

## Placed in

- Multimodal LLMs / vision-language models / [Prompt learning and adaptation of CLIP](https://mlsanity.com/mllm/prompt-learning-and-adaptation-of-clip/)

## Builds on (most cited first, 21 of 21)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Learning to Prompt for Vision-Language Models](https://mlsanity.com/paper/2109.01134/) (2021)
- [EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification](https://mlsanity.com/paper/1709.00029/) (2017)
- [Conditional Prompt Learning for Vision-Language Models](https://mlsanity.com/paper/2203.05557/) (2022)
- [Video-LLaVA: Learning United Visual Representation by Alignment Before Projection](https://mlsanity.com/paper/2311.10122/) (2023)
- [CogVLM: Visual Expert for Pretrained Language Models](https://mlsanity.com/paper/2311.03079/) (2023)
- [Tip-Adapter: Training-free Adaption of CLIP for Few-shot Classification](https://mlsanity.com/paper/2207.09519/) (2022)
- [Synthetic Data from Diffusion Models Improves ImageNet Classification](https://mlsanity.com/paper/2304.08466/) (2023)
- [Is synthetic data from generative models ready for image recognition?](https://mlsanity.com/paper/2210.07574/) (2022)
- [SuS-X: Training-Free Name-Only Transfer of Vision-Language Models](https://mlsanity.com/paper/2211.16198/) (2022)
- [CALIP: Zero-Shot Enhancement of CLIP with Parameter-free Attention](https://mlsanity.com/paper/2209.14169/) (2022)
- [LayoutLLM-T2I: Eliciting Layout Guidance from LLM for Text-to-Image Generation](https://mlsanity.com/paper/2308.05095/) (2023)
- [Improved Zero-Shot Classification by Adapting VLMs with Text Descriptions](https://mlsanity.com/paper/2401.02460/) (2024)
- [Few-Shot Object Detection via Association and DIscrimination](https://mlsanity.com/paper/2111.11656/) (2021)
- [MosaicFusion: Diffusion Models as Data Augmenters for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/2309.13042/) (2023)
- [Online Zero-Shot Classification with CLIP](https://mlsanity.com/paper/2408.13320/) (2024)
- [Interpretable Zero-Shot Learning with Locally-Aligned Vision-Language Model](https://mlsanity.com/paper/2506.23822/) (2025)
- [Discriminative Image Generation with Diffusion Models for Zero-Shot Learning](https://mlsanity.com/paper/2412.17219/) (2024)
- [Contrastive Visual Data Augmentation](https://mlsanity.com/paper/2502.17709/) (2025)

## Abstract (arXiv)

Vision-Language Models (VLMs) have shown strong performance in zero-shot image classification tasks. However, existing methods, including Contrastive Language-Image Pre-training (CLIP), all rely on annotated text-to-image pairs for aligning visual and textual modalities. This dependency introduces substantial cost and accuracy requirement in preparing high-quality datasets. At the same time, processing data from two modes also requires dual-tower encoders for most models, which also hinders their lightweight. To address these limitations, we introduce a ``Contrastive Language-Image Pre-training via Large-Language-Model-based Generation (LGCLIP)" framework. LGCLIP leverages a Large Language Model (LLM) to generate class-specific prompts that guide a diffusion model in synthesizing reference images. Afterwards these generated images serve as visual prototypes, and the visual features of real images are extracted and compared with the visual features of these prototypes to achieve comparative prediction. By optimizing prompt generation through the LLM and employing only a visual encoder, LGCLIP remains lightweight and efficient. Crucially, our framework requires only class labels as input during whole experimental procedure, eliminating the need for manually annotated image-text pairs and extra pre-processing. Experimental results validate the feasibility and efficiency of LGCLIP, demonstrating great performance in zero-shot classification tasks and establishing a novel paradigm for classification.