HALO: Hardware-Aware Quantization with Low Critical-Path-Delay Weights for LLM Acceleration
Rohan Juneja, Shivam Aggarwal, Safeen Huda, Tulika Mitra, Li-Shiuan Peh
Abstract
Quantization is critical for efficiently deploying large language models (LLMs). Yet conventional methods remain hardware-agnostic, limited to bit-width constraints, and do not account for intrinsic circuit characteristics such as the timing behaviors and energy profiles of Multiply-Accumulate (MAC) units. This disconnect from circuit-level behavior limits the ability to exploit available timing margins and energy-saving opportunities, reducing the overall efficiency of deployment on modern accelerators. To address these limitations, we propose HALO, a versatile framework for Hardware-Aware Post-Training Quantization (PTQ). Unlike traditional methods, HALO explicitly incorporates detailed hardware characteristics, including critical-path timing and power consumption, into its quantization approach. HALO strategically selects weights with low critical-path-delays enabling higher operational frequencies and dynamic frequency scaling without disrupting the architecture
BibTeX
@inproceedings{aaai2026_halohardwareawar,
title = {HALO: Hardware-Aware Quantization with Low Critical-Path-Delay Weights for LLM Acceleration},
author = {Rohan Juneja and Shivam Aggarwal and Safeen Huda and Tulika Mitra and Li-Shiuan Peh},
booktitle = {AAAI 2026},
year = {2026}
}