Measuring Accuracy and Energy-to-Solution of Quantum Fine-Tuning of Foundational AI Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Knitter, Oliver, Kim, Sang Hyub, Wurzer, Maximilian, Mei, Jonathan, Girotto, Claudio, Horovitz, Karen, Chen, Chi, Yamada, Masako, Flöther, Frederik F., Roetteler, Martin
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916001602338816
author Knitter, Oliver
Kim, Sang Hyub
Wurzer, Maximilian
Mei, Jonathan
Girotto, Claudio
Horovitz, Karen
Chen, Chi
Yamada, Masako
Flöther, Frederik F.
Roetteler, Martin
author_facet Knitter, Oliver
Kim, Sang Hyub
Wurzer, Maximilian
Mei, Jonathan
Girotto, Claudio
Horovitz, Karen
Chen, Chi
Yamada, Masako
Flöther, Frederik F.
Roetteler, Martin
contents We present an experimental study of energy-to-solution (ETS) of hybrid quantum-classical applications, enabled by direct instrumentation of power consumption of a Forte Enterprise trapped-ion quantum processor. We apply this methodology to a hybrid quantum-classical pipeline for quantum fine-tuning of foundational AI models, and validate the approach end-to-end on quantum hardware. Despite noise and limited qubit counts, the resulting models achieve accuracy competitive with and exceeding classical baselines such as logistic regression and support vector classifiers. Our results show that QPU energy consumption scales approximately linearly with qubit number for shallow circuits, while classical simulation exhibits exponential scaling, indicating a break-even for ETS around 34 qubits. The classification error improvement of the best quantum fine-tuned model over the best classical fine-tuned model considered in this study is around 24%. We further contextualize these findings with comparisons to tensor network methods. This work establishes energy-to-solution as a measurable and scalable metric for evaluating quantum applications and provides experimental evidence of favorable energy-accuracy trade-offs.
format Preprint
id arxiv_https___arxiv_org_abs_2605_02798
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Measuring Accuracy and Energy-to-Solution of Quantum Fine-Tuning of Foundational AI Models
Knitter, Oliver
Kim, Sang Hyub
Wurzer, Maximilian
Mei, Jonathan
Girotto, Claudio
Horovitz, Karen
Chen, Chi
Yamada, Masako
Flöther, Frederik F.
Roetteler, Martin
Quantum Physics
Artificial Intelligence
Emerging Technologies
Machine Learning
We present an experimental study of energy-to-solution (ETS) of hybrid quantum-classical applications, enabled by direct instrumentation of power consumption of a Forte Enterprise trapped-ion quantum processor. We apply this methodology to a hybrid quantum-classical pipeline for quantum fine-tuning of foundational AI models, and validate the approach end-to-end on quantum hardware. Despite noise and limited qubit counts, the resulting models achieve accuracy competitive with and exceeding classical baselines such as logistic regression and support vector classifiers. Our results show that QPU energy consumption scales approximately linearly with qubit number for shallow circuits, while classical simulation exhibits exponential scaling, indicating a break-even for ETS around 34 qubits. The classification error improvement of the best quantum fine-tuned model over the best classical fine-tuned model considered in this study is around 24%. We further contextualize these findings with comparisons to tensor network methods. This work establishes energy-to-solution as a measurable and scalable metric for evaluating quantum applications and provides experimental evidence of favorable energy-accuracy trade-offs.
title Measuring Accuracy and Energy-to-Solution of Quantum Fine-Tuning of Foundational AI Models
topic Quantum Physics
Artificial Intelligence
Emerging Technologies
Machine Learning
url https://arxiv.org/abs/2605.02798