Investigating Data-to-Text Approaches to Achieve Diversity of Generated Marketing Text in the Music Industry
Anum Afzal, Alexandre Mercier, Florian Matthes
Technical University of Munich
Online platforms are increasingly interested in using Data-to-Text technologies to generate content and help their users. Unfortunately, traditional generative methods often fall into repetitive patterns, resulting in monotonous galleries of texts after only a few iterations. To this end, we use a platform designed for musicians and event organizers to create machine-generated band descriptions as our use case. In this paper, we investigate LLM-based data-to-text approaches to automatically generate marketing texts that are both of sufficient quality and diverse enough for broad adoption. We leverage Language Models such as T5, GPT-3.5, GPT-4, and LLaMa2 in conjunction with fine-tuning, few-shot, and zero-shot approaches to set a baseline for diverse marketing texts. We also introduce a metric to evaluate the diversity of a set of texts and additionally use G-eval as a quality metric. We propose solutions both at the prompting stage and the decoding stage and evaluate their impact on diversity. This research extends its relevance beyond the music industry, proving beneficial in various fields where repetitive automated content generation is prevalent.