Quant-Trim in Practice: Improved Cross-Platform Low-Bit Deployment on Edge NPUs

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Dhahri, Rayen, Urban, Steffen
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915627206180864
author Dhahri, Rayen
Urban, Steffen
author_facet Dhahri, Rayen
Urban, Steffen
contents Specialized edge accelerators rely on low-bit quantization, but vendor compilers differ in scaling, clipping, and kernel support, often as black boxes. The same floating-point (FP) checkpoint can therefore yield inconsistent accuracy across backends, forcing practitioners to tweak flags or refactor models to vendor-friendly operator subsets. We introduce Quant-Trim, a training-phase method that produces a hardware-neutral checkpoint robust to backend and precision choices. It combines progressive fake quantization to align training with the deployed integer grid and reverse pruning to tame outlier-driven scale inflation while preserving learnability. Quant-Trim is agnostic to quantization schemes (symmetric/asymmetric, per-tensor/per-channel, INT8/INT4) and requires no vendor-specific graph changes. Across models and tasks, it narrows the FP-to-low-bit gap, reduces dependence on compiler heuristics/calibration, and avoids per-backend retraining. We report accuracy and edge metrics latency, throughput, energy per inference, and cost under static/dynamic activation scaling and varying operator coverage.
format Preprint
id arxiv_https___arxiv_org_abs_2511_15300
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Quant-Trim in Practice: Improved Cross-Platform Low-Bit Deployment on Edge NPUs
Dhahri, Rayen
Urban, Steffen
Machine Learning
Specialized edge accelerators rely on low-bit quantization, but vendor compilers differ in scaling, clipping, and kernel support, often as black boxes. The same floating-point (FP) checkpoint can therefore yield inconsistent accuracy across backends, forcing practitioners to tweak flags or refactor models to vendor-friendly operator subsets. We introduce Quant-Trim, a training-phase method that produces a hardware-neutral checkpoint robust to backend and precision choices. It combines progressive fake quantization to align training with the deployed integer grid and reverse pruning to tame outlier-driven scale inflation while preserving learnability. Quant-Trim is agnostic to quantization schemes (symmetric/asymmetric, per-tensor/per-channel, INT8/INT4) and requires no vendor-specific graph changes. Across models and tasks, it narrows the FP-to-low-bit gap, reduces dependence on compiler heuristics/calibration, and avoids per-backend retraining. We report accuracy and edge metrics latency, throughput, energy per inference, and cost under static/dynamic activation scaling and varying operator coverage.
title Quant-Trim in Practice: Improved Cross-Platform Low-Bit Deployment on Edge NPUs
topic Machine Learning
url https://arxiv.org/abs/2511.15300