Saved in:
Bibliographic Details
Main Authors: Shin, Juncheol, Seok, Minsang, Kim, Seonggon, Park, Eunhyeok
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.23651
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916766568939520
author Shin, Juncheol
Seok, Minsang
Kim, Seonggon
Park, Eunhyeok
author_facet Shin, Juncheol
Seok, Minsang
Kim, Seonggon
Park, Eunhyeok
contents Model merging has emerged as a powerful technique for combining task-specific weights, achieving superior performance in multi-target domain adaptation. However, when applied to practical scenarios, such as quantized models, new challenges arise. In practical scenarios, quantization is often applied to target-specific data, but this process restricts the domain of interest and introduces discretization effects, making model merging highly non-trivial. In this study, we analyze the impact of quantization on model merging through the lens of error barriers. Leveraging these insights, we propose a novel post-training quantization, HDRQ - Hessian and distant regularizing quantization - that is designed to consider model merging for multi-target domain adaptation. Our approach ensures that the quantization process incurs minimal deviation from the source pre-trained model while flattening the loss surface to facilitate smooth model merging. To our knowledge, this is the first study on this challenge, and extensive experiments confirm its effectiveness.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23651
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Merge-Friendly Post-Training Quantization for Multi-Target Domain Adaptation
Shin, Juncheol
Seok, Minsang
Kim, Seonggon
Park, Eunhyeok
Machine Learning
Computer Vision and Pattern Recognition
Model merging has emerged as a powerful technique for combining task-specific weights, achieving superior performance in multi-target domain adaptation. However, when applied to practical scenarios, such as quantized models, new challenges arise. In practical scenarios, quantization is often applied to target-specific data, but this process restricts the domain of interest and introduces discretization effects, making model merging highly non-trivial. In this study, we analyze the impact of quantization on model merging through the lens of error barriers. Leveraging these insights, we propose a novel post-training quantization, HDRQ - Hessian and distant regularizing quantization - that is designed to consider model merging for multi-target domain adaptation. Our approach ensures that the quantization process incurs minimal deviation from the source pre-trained model while flattening the loss surface to facilitate smooth model merging. To our knowledge, this is the first study on this challenge, and extensive experiments confirm its effectiveness.
title Merge-Friendly Post-Training Quantization for Multi-Target Domain Adaptation
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.23651