# Post Training Quantization for Efficient Dataset Condensation

arXiv: https://arxiv.org/abs/2603.13346 · HTML: https://mlsanity.com/paper/2603.13346/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 29 references that are in the map.
- First posted on arXiv 7 Mar 2026.

## Builds on (most cited first, 20 of 20)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Learned Step Size Quantization](https://mlsanity.com/paper/1902.08153/) (2019)
- [Dataset Condensation with Gradient Matching](https://mlsanity.com/paper/2006.05929/) (2020)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [Dataset Distillation by Matching Training Trajectories](https://mlsanity.com/paper/2203.11932/) (2022)
- [Dataset Condensation with Differentiable Siamese Augmentation](https://mlsanity.com/paper/2102.08259/) (2021)
- [Dataset Distillation](https://mlsanity.com/paper/1811.10959/) (2018)
- [Dataset Condensation via Efficient Synthetic-Data Parameterization](https://mlsanity.com/paper/2205.14959/) (2022)
- [Scaling Up Dataset Distillation to ImageNet-1K with Constant Memory](https://mlsanity.com/paper/2211.10586/) (2022)
- [PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization](https://mlsanity.com/paper/2111.12293/) (2021)
- [Improving the Robustness of Deep Neural Networks via Stability Training](https://mlsanity.com/paper/1604.04326/) (2016)
- [Improved Distribution Matching for Dataset Condensation](https://mlsanity.com/paper/2307.09742/) (2023)
- [Privacy for Free: How does Dataset Condensation Help Privacy?](https://mlsanity.com/paper/2206.00240/) (2022)
- [Towards Lossless Dataset Distillation via Difficulty-Aligned Trajectory Matching](https://mlsanity.com/paper/2310.05773/) (2023)
- [ZeroQuant-V2: Exploring Post-training Quantization in LLMs from Comprehensive Study to Low Rank Compensation](https://mlsanity.com/paper/2303.08302/) (2023)
- [Examining the Impact of Blur on Recognition by Convolutional Networks](https://mlsanity.com/paper/1611.05760/) (2016)
- [Summarizing Stream Data for Memory-Constrained Online Continual Learning](https://mlsanity.com/paper/2305.16645/) (2023)
- [Color-Oriented Redundancy Reduction in Dataset Distillation](https://mlsanity.com/paper/2411.11329/) (2024)

## Built on by (most cited first, 2 of 2)

- [Closed-Form Linear-Probe Dataset Distillation for Pre-trained Vision Models](https://mlsanity.com/paper/2605.07194/) (2026)
- [Long-Tailed 3D Point Cloud Dataset Distillation](https://mlsanity.com/paper/2607.26763/) (2026)

## Abstract (arXiv)

Dataset Condensation (DC) distills knowledge from large datasets into smaller ones, accelerating training and reducing storage requirements. However, despite notable progress, prior methods have largely overlooked the potential of quantization for further reducing storage costs. In this paper, we take the first step to explore post-training quantization in dataset condensation, demonstrating its effectiveness in reducing storage size while maintaining representation quality without requiring expensive training cost. However, we find that at extremely low bit-widths (e.g., 2-bit), conventional quantization leads to substantial degradation in representation quality, negatively impacting the networks trained on these data. To address this, we propose a novel patch-based post-training quantization approach that ensures localized quantization with minimal loss of information. To reduce the overhead of quantization parameters, especially for small patch sizes, we employ quantization-aware clustering to identify similar patches and subsequently aggregate them for efficient quantization. Furthermore, we introduce a refinement module to align the distribution between original images and their dequantized counterparts, compensating for quantization errors. Our method is a plug-and-play framework that can be applied to synthetic images generated by various DC methods. Extensive experiments across diverse benchmarks including CIFAR-10/100, Tiny ImageNet, and ImageNet subsets demonstrate that our method consistently outperforms prior works under the same storage constraints. Notably, our method nearly doubles the test accuracy of existing methods at extreme compression regimes (e.g., 26.0% rightarrow 54.1% for DM at IPC=1), while operating directly on 2-bit images without additional distillation.