Saved in:
Bibliographic Details
Main Authors: Zhang, Shi-Xin, Chen, Yu-Qin
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.29557
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917543254425600
author Zhang, Shi-Xin
Chen, Yu-Qin
author_facet Zhang, Shi-Xin
Chen, Yu-Qin
contents Machine learning models can inherit hidden behavioral traits through innocuous public interfaces, a phenomenon known as subliminal learning. Here we extend this framework to quantum models and study two distillation pathways: an auxiliary channel on random inputs and a restricted task channel in which the student matches a public supervised output while the hidden behavior resides on a disjoint task. Both classical and quantum neural networks (QNNs) exhibit efficient auxiliary-channel subliminal learning, but the task channel shows strong architecture dependence. Classical neural networks transmit little hidden-task information through the public-task interface, whereas QNNs retain most of the hidden-task signal. We show that a unified geometric picture explains both regimes: transmission is controlled by the teacher drift magnitude together with the fraction of hidden-task-relevant drift that remains visible through the public interface. These results identify a concrete security concern for quantum model supply chains and suggest a controlled route for hidden-information transfer in quantum information processing.
format Preprint
id arxiv_https___arxiv_org_abs_2605_29557
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Quantum Subliminal Learning
Zhang, Shi-Xin
Chen, Yu-Qin
Quantum Physics
Machine learning models can inherit hidden behavioral traits through innocuous public interfaces, a phenomenon known as subliminal learning. Here we extend this framework to quantum models and study two distillation pathways: an auxiliary channel on random inputs and a restricted task channel in which the student matches a public supervised output while the hidden behavior resides on a disjoint task. Both classical and quantum neural networks (QNNs) exhibit efficient auxiliary-channel subliminal learning, but the task channel shows strong architecture dependence. Classical neural networks transmit little hidden-task information through the public-task interface, whereas QNNs retain most of the hidden-task signal. We show that a unified geometric picture explains both regimes: transmission is controlled by the teacher drift magnitude together with the fraction of hidden-task-relevant drift that remains visible through the public interface. These results identify a concrete security concern for quantum model supply chains and suggest a controlled route for hidden-information transfer in quantum information processing.
title Quantum Subliminal Learning
topic Quantum Physics
url https://arxiv.org/abs/2605.29557