Saved in:
Bibliographic Details
Main Authors: Kazemi, Alireza, Rezvani, Helia, Baktashmotlagh, Mahsa
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.20121
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908342613442560
author Kazemi, Alireza
Rezvani, Helia
Baktashmotlagh, Mahsa
author_facet Kazemi, Alireza
Rezvani, Helia
Baktashmotlagh, Mahsa
contents Transferability scores aim to quantify how well a model trained on one domain generalizes to a target domain. Despite numerous methods proposed for measuring transferability, their reliability and practical usefulness remain inconclusive, often due to differing experimental setups, datasets, and assumptions. In this paper, we introduce a comprehensive benchmarking framework designed to systematically evaluate transferability scores across diverse settings. Through extensive experiments, we observe variations in how different metrics perform under various scenarios, suggesting that current evaluation practices may not fully capture each method's strengths and limitations. Our findings underscore the value of standardized assessment protocols, paving the way for more reliable transferability measures and better-informed model selection in cross-domain applications. Additionally, we achieved a 3.5\% improvement using our proposed metric for the head-training fine-tuning experimental setup. Our code is available in this repository: https://github.com/alizkzm/pert_robust_platform.
format Preprint
id arxiv_https___arxiv_org_abs_2504_20121
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Benchmarking Transferability: A Framework for Fair and Robust Evaluation
Kazemi, Alireza
Rezvani, Helia
Baktashmotlagh, Mahsa
Machine Learning
Transferability scores aim to quantify how well a model trained on one domain generalizes to a target domain. Despite numerous methods proposed for measuring transferability, their reliability and practical usefulness remain inconclusive, often due to differing experimental setups, datasets, and assumptions. In this paper, we introduce a comprehensive benchmarking framework designed to systematically evaluate transferability scores across diverse settings. Through extensive experiments, we observe variations in how different metrics perform under various scenarios, suggesting that current evaluation practices may not fully capture each method's strengths and limitations. Our findings underscore the value of standardized assessment protocols, paving the way for more reliable transferability measures and better-informed model selection in cross-domain applications. Additionally, we achieved a 3.5\% improvement using our proposed metric for the head-training fine-tuning experimental setup. Our code is available in this repository: https://github.com/alizkzm/pert_robust_platform.
title Benchmarking Transferability: A Framework for Fair and Robust Evaluation
topic Machine Learning
url https://arxiv.org/abs/2504.20121