• Authors: Thomas Winters, Pieter Delobelle
• Publication Date: 2021-09-15
• Publication Venue: Proceedings of the Twelfth International Conference on Computational Creativity
• Abstract: Large pre-trained transformer-based language models have revolutionized the field of natural language processing in recent years. While BERT-like models perform exceptionally well for analytical tasks such as classification and regression, their text generation capabilities are usually limited to predicting tokens within a given context. In this paper, we introduce GALMET, a model that generates text by using genetic algorithms with BERT-like language models for evolving text. We use GALMET with the RoBERTa language model to automatically evolve real headlines into more satirical headlines. This is achieved by adapting the masked language head to the headlines domain for the mutation operator and finetuning a regression head to distinguish headlines from satire for the fitness function. We evaluated our system by comparing generated satirical headlines against human-edited headlines and just the fine-tuned masked language head. We found that while humans generally outperform the model, generations by GALMET are also often preferred over human-edited headlines. However, we also found that only using the fine-tuned masked language model gives slightly preferred satire due to generating more readable sentences. GALMET is thus a first step towards a new way of creating text generators using masked language models by transforming text guided by scores from another language model.

Citation

APA

Harvard

Vancouver

BibTeX