让边缘视觉模型更好支持低比特量化部署。
SCULPT: Training Edge Vision Models for Post-Training Quantization Readiness
- 训练时用统计裁剪和均匀损失抑制激活分布偏斜。
- 在INT8及以下低比特下仍保持高精度，无需额外修复步骤。
- 适合希望简化部署流程的模型开发者使用。
边缘视觉模型在资源受限设备上部署困难，低比特后训练量化（PTQ）因此备受关注。标准FP32训练常导致激活分布呈重尾状，极端值使量化不稳定：保留全范围会浪费量化桶，过度截断又造成信息丢失。现有方法多依赖量化感知训练（QAT），增加训练复杂度且与位宽耦合；或采用训练后修复的高级PTQ流程。本文提出SCULPT（Statistical Clipping and Uniform Loss for Post-Training），一种在普通FP32微调期间提升PTQ就绪性的训练方法。SCULPT结合拓扑感知激活正则项以抑制量化不友好偏度与峰度，并引入基于分位数的稳定裁剪机制，学习适用于部署的激活边界。不同于QAT，SCULPT不模拟量化过程；也不同于事后异常值修复型PTQ方法，无需运行时激活变换。学习到的裁剪边界可直接导出至标准PTQ流程，支持包括INT8及更低比特如W4A8在内的低比特部署。
原文摘要 · Abstract (English)
Edge vision models are difficult to deploy on resource-constrained hardware, making low-bit post-training quantization (PTQ) attractive. In practice, standard FP32 training often produces heavy-tailed activation distributions whose outliers destabilize activation quantization: preserving the full range wastes quantization bins on rare extremes, while aggressive clipping causes information loss. Existing solutions typically rely on quantization-aware training (QAT), which adds training complexity and bit-width coupling, or advanced PTQ procedures that repair the model after training. We present SCULPT (Statistical Clipping and Uniform Loss for Post-Training), a training-time method that improves PTQ readiness during ordinary FP32 fine-tuning. SCULPT combines a topology-aware activation regularizer that suppresses quantization-hostile skewness and kurtosis with a stable percentile-based clipping mechanism that learns deployment-ready activation bounds. Unlike QAT, SCULPT does not simulate quantization during optimization; unlike post hoc outlier-repair PTQ methods, it does not require runtime activation transformations. The learned clipping bounds can be exported directly into a standard PTQ workflow for low-bit deployment, including INT8 and lower-bit settings such as W4A8.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。