# RAMP: Robust Adaptive Mixed-Precision Quantization for Edge CPU Vision Models

arXiv: https://arxiv.org/abs/2609.28262 · HTML: https://mlsanity.com/paper/2609.28262/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 40 references that are in the map.
- First posted on arXiv 23 Sep 2026.

## Builds on (most cited first, 28 of 28)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers](https://mlsanity.com/paper/2206.01861/) (2022)
- [Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues](https://mlsanity.com/paper/2007.09355/) (2020)
- [Leveraging Frequency Analysis for Deep Fake Image Recognition](https://mlsanity.com/paper/2003.08685/) (2020)
- [HAQ: Hardware-Aware Automated Quantization with Mixed Precision](https://mlsanity.com/paper/1811.08886/) (2018)
- [Single Path One-Shot Neural Architecture Search with Uniform Sampling](https://mlsanity.com/paper/1904.00420/) (2019)
- [HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural Networks](https://mlsanity.com/paper/1911.03852/) (2019)
- [TinyViT: Fast Pretraining Distillation for Small Vision Transformers](https://mlsanity.com/paper/2207.10666/) (2022)
- [A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and Recommendations](https://mlsanity.com/paper/2308.06767/) (2023)
- [Low-bit Quantization of Neural Networks for Efficient Inference](https://mlsanity.com/paper/1902.06822/) (2019)
- [Mixed Precision Quantization of ConvNets via Differentiable Neural Architecture Search](https://mlsanity.com/paper/1812.00090/) (2018)
- [APQ: Joint Search for Network Architecture, Pruning and Quantization Policy](https://mlsanity.com/paper/2006.08509/) (2020)
- [Rethinking Differentiable Search for Mixed-Precision Neural Networks](https://mlsanity.com/paper/2004.05795/) (2020)
- [AutoQ: Automated Kernel-Wise Neural Network Quantization](https://mlsanity.com/paper/1902.05690/) (2019)
- [FBGEMM: Enabling High-Performance Low-Precision Deep Learning Inference](https://mlsanity.com/paper/2101.05615/) (2021)
- [A Practical Mixed Precision Algorithm for Post-Training Quantization](https://mlsanity.com/paper/2302.05397/) (2023)
- [EMQ: Evolving Training-free Proxies for Automated Mixed Precision Quantization](https://mlsanity.com/paper/2307.10554/) (2023)
- [BatchQuant: Quantized-for-all Architecture Search with Robust Quantizer](https://mlsanity.com/paper/2105.08952/) (2021)
- [AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models](https://mlsanity.com/paper/2509.12019/) (2025)
- [Channel-Wise Mixed-Precision Quantization for Large Language Models](https://mlsanity.com/paper/2410.13056/) (2024)
- [Mixed-Precision Quantization for Deep Vision Models with Integer Quadratic Programming](https://mlsanity.com/paper/2307.05657/) (2023)
- [Mix-QSAM: Mixed-Precision Quantization of the Segment Anything Model](https://mlsanity.com/paper/2505.04861/) (2025)
- [InfoQ: Mixed-Precision Quantization via Global Information Flow](https://mlsanity.com/paper/2508.04753/) (2025)
- [Present and Future Generalization of Synthetic Image Detectors](https://mlsanity.com/paper/2409.14128/) (2024)
- [DQT: Dynamic Quantization Training via Dequantization-Free Nested Integer Arithmetic](https://mlsanity.com/paper/2508.09176/) (2025)
- [SigmaQuant: Hardware-Aware Heterogeneous Quantization Method for Edge DNN Inference](https://mlsanity.com/paper/2602.22136/) (2026)

## Abstract (arXiv)

Deploying deep learning models on edge CPUs is bottlenecked by computational and memory constraints. Mixed-precision quantization promises to reduce inference latency while preserving accuracy. However, quantization affects different layer types in inconsistent ways, so identifying where accuracy loss is minimized and latency reduction is maximized is critical, as the effect accumulates over a full deployment into substantial savings or unacceptable task degradation. Such identification relies on sensitivity metrics, proxies that estimate layer-wise degradation without evaluating the task accuracy of every candidate policy. Nevertheless, widely used metrics fail systematically on modern architectures. We present a systematic empirical study of 13 sensitivity metrics for layer-wise INT8 quantization across four distinctly different neural networks, and validate the resulting policies on two ARM64 platforms. Gradient-based sensitivity methods fail on 4 out of 8 model-hardware configurations and weight-based statistics on 2. In contrast, the Jensen-Shannon Divergence achieves zero catastrophic failures, reliably isolating the layers that cannot be safely quantized. A sensitivity metric alone does not define a policy, and the fixed thresholds typically used for that step are fragile over the highly skewed distributions of modern architectures. We address this with K-Means clustering, achieving near-lossless accuracy and a mean speed-up of 1.81× over the full-precision model. Finally, we reveal that excluding from quantization the layers whose speed-up is negligible, regardless of their sensitivity, can be counterproductive, as it induces computational graph fragmentation and disables operator fusion. Our results yield concrete allocation policies for practitioners and researchers deploying quantized vision models on heterogeneous edge CPUs, without GPU access or gradient computation.