Abstract
This publication describes a novel system and method for evaluating artificial intelligence (AI)-generated media, focusing on prompt adherence, technical quality, and brand alignment. The framework addresses challenges in current evaluation methods by employing dynamic, multi-component rubric generation and a multimodal, rubric-based evaluation and scoring process. This approach provides a scalable, reliable, and interpretable solution for comprehensively assessing generative media.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Fekry, Mohab; Koch, Cornelius; and Stellwag, Luka, "Rubric-Based Evaluation for On-Brand Generative Media", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/8935
COinS