Salvato in:
Dettagli Bibliografici
Autori principali: Sarna, Neeraj, Li, Yuanyuan, von Gablenz, Michael
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2512.15442
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918252430491648
author Sarna, Neeraj
Li, Yuanyuan
von Gablenz, Michael
author_facet Sarna, Neeraj
Li, Yuanyuan
von Gablenz, Michael
contents Large scale text-to-image generation models can memorize and reproduce their training dataset. Since the training dataset often contains copyrighted material, reproduction of training dataset poses a copyright infringement risk, which could result in legal liabilities and financial losses for both the AI user and the developer. The current works explores the potential of chain-of-thought and task instruction prompting in reducing copyrighted content generation. To this end, we present a formulation that combines these two techniques with two other copyright mitigation strategies: a) negative prompting, and b) prompt re-writing. We study the generated images in terms their similarity to a copyrighted image and their relevance of the user input. We present numerical experiments on a variety of models and provide insights on the effectiveness of the aforementioned techniques for varying model complexity.
format Preprint
id arxiv_https___arxiv_org_abs_2512_15442
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Copyright Infringement Risk Reduction via Chain-of-Thought and Task Instruction Prompting
Sarna, Neeraj
Li, Yuanyuan
von Gablenz, Michael
Machine Learning
Large scale text-to-image generation models can memorize and reproduce their training dataset. Since the training dataset often contains copyrighted material, reproduction of training dataset poses a copyright infringement risk, which could result in legal liabilities and financial losses for both the AI user and the developer. The current works explores the potential of chain-of-thought and task instruction prompting in reducing copyrighted content generation. To this end, we present a formulation that combines these two techniques with two other copyright mitigation strategies: a) negative prompting, and b) prompt re-writing. We study the generated images in terms their similarity to a copyrighted image and their relevance of the user input. We present numerical experiments on a variety of models and provide insights on the effectiveness of the aforementioned techniques for varying model complexity.
title Copyright Infringement Risk Reduction via Chain-of-Thought and Task Instruction Prompting
topic Machine Learning
url https://arxiv.org/abs/2512.15442