Copyright Infringement Risk Reduction via Chain-of-Thought and Task Instruction Prompting
This addresses copyright risks for AI users and developers, but it is incremental as it builds on existing mitigation strategies.
The paper tackles the problem of copyright infringement risk in text-to-image generation models by proposing a method combining chain-of-thought and task instruction prompting with negative prompting and prompt re-writing, resulting in reduced similarity to copyrighted images while maintaining relevance to user inputs as shown in numerical experiments across various models.
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.