# Outlier-Aware Post-Training Quantization for Image Super-Resolution

arXiv: https://arxiv.org/abs/2511.00682 · HTML: https://mlsanity.com/paper/2511.00682/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in KV cache quantization, one of 12 directions in Efficient LLM inference and serving.
- Builds on 32 of its 54 references that are in the map.
- First posted on arXiv 1 Nov 2025.

## Placed in

- Efficient LLM inference and serving / [KV cache quantization](https://mlsanity.com/inference/kv-cache-quantization/)

## Builds on (most cited first, 30 of 32)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network](https://mlsanity.com/paper/1609.04802/) (2016)
- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [Enhanced Deep Residual Networks for Single Image Super-Resolution](https://mlsanity.com/paper/1707.02921/) (2017)
- [Image Super-Resolution Using Very Deep Residual Channel Attention Networks](https://mlsanity.com/paper/1807.02758/) (2018)
- [Image Super-Resolution Using Deep Convolutional Networks](https://mlsanity.com/paper/1501.00092/) (2014)
- [Residual Dense Network for Image Super-Resolution](https://mlsanity.com/paper/1802.08797/) (2018)
- [Post-Training Quantization for Vision Transformer](https://mlsanity.com/paper/2106.14156/) (2021)
- [Post-training Quantization on Diffusion Models](https://mlsanity.com/paper/2211.15736/) (2022)
- [Fast, Accurate, and Lightweight Super-Resolution with Cascading Residual Network](https://mlsanity.com/paper/1803.08664/) (2018)
- [PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization](https://mlsanity.com/paper/2111.12293/) (2021)
- [Efficient Long-Range Attention Network for Image Super-resolution](https://mlsanity.com/paper/2203.06697/) (2022)
- [ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic](https://mlsanity.com/paper/2103.04039/) (2021)
- [Post-Training Piecewise Linear Quantization for Deep Neural Networks](https://mlsanity.com/paper/2002.00104/) (2020)
- [Spatially-Adaptive Feature Modulation for Efficient Image Super-Resolution](https://mlsanity.com/paper/2302.13800/) (2023)
- [PAMS: Quantized Super-Resolution via Parameterized Max Scale](https://mlsanity.com/paper/2011.04212/) (2020)
- [AdaBin: Improving Binary Neural Networks with Adaptive Binary Sets](https://mlsanity.com/paper/2208.08084/) (2022)
- [DAQ: Channel-Wise Distribution-Aware Quantization for Deep Image Super-Resolution Networks](https://mlsanity.com/paper/2012.11230/) (2020)
- [CADyQ: Content-Aware Dynamic Quantization for Image Super-Resolution](https://mlsanity.com/paper/2207.10345/) (2022)
- [Once Quantization-Aware Training: High Performance Extremely Low-bit Architecture Search](https://mlsanity.com/paper/2010.04354/) (2020)
- [Dynamic Dual Trainable Bounds for Ultra-low Precision Super-Resolution Networks](https://mlsanity.com/paper/2203.03844/) (2022)
- [Bridging the Accuracy Gap for 2-bit Quantized Neural Networks (QNN)](https://mlsanity.com/paper/1807.06964/) (2018)
- [Fully Quantized Image Super-Resolution Networks](https://mlsanity.com/paper/2011.14265/) (2020)
- [DIPNet: Efficiency Distillation and Iterative Pruning for Image Super-Resolution](https://mlsanity.com/paper/2304.07018/) (2023)
- [Edge-enhanced Feature Distillation Network for Efficient Super-Resolution](https://mlsanity.com/paper/2204.08759/) (2022)
- [CABM: Content-Aware Bit Mapping for Single Image Super-Resolution Network with Large Input](https://mlsanity.com/paper/2304.06454/) (2023)
- [Structured Sparsity Learning for Efficient Video Super-Resolution](https://mlsanity.com/paper/2206.07687/) (2022)
- [AdaBM: On-the-Fly Adaptive Bit Mapping for Image Super-Resolution](https://mlsanity.com/paper/2404.03296/) (2024)
- [Differentiable Neural Architecture Search for Extremely Lightweight Image Super-Resolution](https://mlsanity.com/paper/2105.03939/) (2021)
- [Overcoming Distribution Mismatch in Quantizing Image Super-Resolution Networks](https://mlsanity.com/paper/2307.13337/) (2023)

## Built on by (most cited first, 4 of 4)

- [Restore-R1: Efficient Image Restoration Agents via Reinforcement Learning with Multimodal LLM Perceptual Feedback](https://mlsanity.com/paper/2512.18599/) (2025)
- [Seeing Through Words: Controlling Visual Retrieval Quality with Language Models](https://mlsanity.com/paper/2602.21175/) (2026)
- [QuantSR+: Pushing the Limit of Quantized Image Super-Resolution Networks](https://mlsanity.com/paper/2605.22351/) (2026)
- [Hardware-Friendly Token-Group Activation Quantization for Low-Bit Mamba Super-Resolution](https://mlsanity.com/paper/2610.16261/) (2026)

## Abstract (arXiv)

Quantization techniques, including quantization-aware training (QAT) and post-training quantization (PTQ), have become essential for inference acceleration of image super-resolution (SR) networks. Compared to QAT, PTQ has garnered significant attention as it eliminates the need for ground truth and model retraining. However, existing PTQ methods for SR often fail to achieve satisfactory performance as they overlook the impact of outliers in activation. Our empirical analysis reveals that these prevalent activation outliers are strongly correlated with image color information, and directly removing them leads to significant performance degradation. Motivated by this, we propose a dual-region quantization strategy that partitions activations into an outlier region and a dense region, applying uniform quantization to each region independently to better balance bit-width allocation. Furthermore, we observe that different network layers exhibit varying sensitivities to quantization, leading to different levels of performance degradation. To address this, we introduce sensitivity-aware finetuning that encourages the model to focus more on highly sensitive layers, further enhancing quantization performance. Extensive experiments demonstrate that our method outperforms existing PTQ approaches across various SR networks and datasets, while achieving performance comparable to QAT methods in most scenarios with at least a 75 speedup.