Compares poem generation effectiveness of Bi-LSTM, Vicuna-1B, and GPT-2, indicating model performance.
Recent advancements in artificial intelligence have significantly impacted various sectors, including the creative domain of poetry generation. This paper examines the effectiveness of neural architectures, specifically Bi-directional Long Short-Term Memory (Bi-LSTM), Vicuna-1B, and GPT-2, in gener-ating poetry. By initiating these models with detailed prompts encompassing genre and theme, we aim to explore their capacity for producing poetry that resonates with human emotion and stylistic precision. Our methodology involves comparing the poems generated by these models against a curated dataset of existing poems, employing BLEU and ROUGE scores for a comprehensive evaluation. This study seeks to shed light on the potential of each model in the field of computational creativity, offering insights into their applicability in poetry generation.
No takes yet. Share an insight, caveat, or question.
G et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: