# Hard Sample Matters a Lot in Zero-Shot Quantization

arXiv: https://arxiv.org/abs/2303.13826 · HTML: https://mlsanity.com/paper/2303.13826/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- Builds on 50 of its 55 references that are in the map.
- First posted on arXiv 24 Mar 2023.

## Builds on (most cited first, 30 of 50)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://mlsanity.com/paper/1704.04861/) (2017)
- [Fast R-CNN](https://mlsanity.com/paper/1504.08083/) (2015)
- [Rich feature hierarchies for accurate object detection and semantic segmentation](https://mlsanity.com/paper/1311.2524/) (2013)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [How transferable are features in deep neural networks?](https://mlsanity.com/paper/1411.1792/) (2014)
- [Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer](https://mlsanity.com/paper/1612.03928/) (2016)
- [Learned Step Size Quantization](https://mlsanity.com/paper/1902.08153/) (2019)
- [BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstruction](https://mlsanity.com/paper/2102.05426/) (2021)
- [PACT: Parameterized Clipping Activation for Quantized Neural Networks](https://mlsanity.com/paper/1805.06085/) (2018)
- [Data-Free Quantization Through Weight Equalization and Bias Correction](https://mlsanity.com/paper/1906.04721/) (2019)
- [PyHessian: Neural Networks Through the Lens of the Hessian](https://mlsanity.com/paper/1912.07145/) (2019)
- [ZeroQ: A Novel Zero Shot Quantization Framework](https://mlsanity.com/paper/2001.00281/) (2020)
- [A Survey of Embodied AI: From Simulators to Research Tasks](https://mlsanity.com/paper/2103.04918/) (2021)
- [Libra R-CNN: Towards Balanced Learning for Object Detection](https://mlsanity.com/paper/1904.02701/) (2019)
- [Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networks](https://mlsanity.com/paper/1908.05033/) (2019)
- [Forward and Backward Information Retention for Accurate Binary Neural Networks](https://mlsanity.com/paper/1909.10788/) (2019)
- [Learning to Quantize Deep Networks by Optimizing Quantization Intervals with Task Loss](https://mlsanity.com/paper/1808.05779/) (2018)
- [Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networks](https://mlsanity.com/paper/1909.13144/) (2019)
- [Federated Class-Incremental Learning](https://mlsanity.com/paper/2203.11473/) (2022)
- [Gradient Harmonized Single-stage Detector](https://mlsanity.com/paper/1811.05181/) (2018)
- [Training Binary Neural Networks with Real-to-Binary Convolutions](https://mlsanity.com/paper/2003.11535/) (2020)
- [Is Out-of-Distribution Detection Learnable?](https://mlsanity.com/paper/2210.14707/) (2022)
- [Generative Low-bitwidth Data Free Quantization](https://mlsanity.com/paper/2003.03603/) (2020)

## Built on by (most cited first, 15 of 15)

- [Enhancing One-Shot Federated Learning Through Data and Ensemble Co-Boosting](https://mlsanity.com/paper/2402.15070/) (2024)
- [EDA-DM: Enhanced Distribution Alignment for Post-Training Quantization of Diffusion Models](https://mlsanity.com/paper/2401.04585/) (2024)
- [A Comprehensive Survey of Convolutions in Deep Learning: Applications, Challenges, and Future Trends](https://mlsanity.com/paper/2402.15490/) (2024)
- [Self-calibration for Language Model Quantization and Pruning](https://mlsanity.com/paper/2410.17170/) (2024)
- [Low-bit Model Quantization for Deep Neural Networks: A Survey](https://mlsanity.com/paper/2505.05530/) (2025)
- [Zero-shot Quantization: A Comprehensive Survey](https://mlsanity.com/paper/2505.09188/) (2025)
- [Sharpness-Aware Data Generation for Zero-shot Quantization](https://mlsanity.com/paper/2510.07018/) (2025)
- [Robustness-Guided Image Synthesis for Data-Free Quantization](https://mlsanity.com/paper/2310.03661/) (2023)
- [Semantic Alignment and Reinforcement for Data-Free Quantization of Vision Transformers](https://mlsanity.com/paper/2412.16553/) (2024)
- [Enhancing Quantization-Aware Training on Edge Devices via Relative Entropy Coreset Selection and Cascaded Layer Correction](https://mlsanity.com/paper/2507.17768/) (2025)
- [Dynamic Shuffle: An Efficient Channel Mixture Method](https://mlsanity.com/paper/2310.02776/) (2023)
- [Infrared Domain Adaptation with Zero-Shot Quantization](https://mlsanity.com/paper/2408.13925/) (2024)
- [Advanced Knowledge Transfer: Refined Feature Distillation for Zero-Shot Quantization in Edge Computing](https://mlsanity.com/paper/2412.19125/) (2024)
- [SynQ: Accurate Zero-shot Quantization by Synthesis-aware Fine-tuning](https://mlsanity.com/paper/2603.18423/) (2026)
- [Close Shortcut Wins Long: Seeking Diverse and Stable Generators for Data-Free Knowledge Distillation](https://mlsanity.com/paper/2608.22003/) (2026)

## Abstract (arXiv)

Zero-shot quantization (ZSQ) is promising for compressing and accelerating deep neural networks when the data for training full-precision models are inaccessible. In ZSQ, network quantization is performed using synthetic samples, thus, the performance of quantized models depends heavily on the quality of synthetic samples. Nonetheless, we find that the synthetic samples constructed in existing ZSQ methods can be easily fitted by models. Accordingly, quantized models obtained by these methods suffer from significant performance degradation on hard samples. To address this issue, we propose HArd sample Synthesizing and Training (HAST). Specifically, HAST pays more attention to hard samples when synthesizing samples and makes synthetic samples hard to fit when training quantized models. HAST aligns features extracted by full-precision and quantized models to ensure the similarity between features extracted by these two models. Extensive experiments show that HAST significantly outperforms existing ZSQ methods, achieving performance comparable to models that are quantized with real data.