The field of computational literature and natural language processing is witnessing significant advancements, particularly in the areas of computational poetry generation, metaphor and analogy extraction, and the annotation of literary texts. A notable trend is the increasing reliance on large language models (LLMs) to tackle complex tasks that require high-level reasoning and language understanding. These models are being employed to generate poems with enhanced structural and rhythmic possibilities, extract metaphoric analogies from literary texts, and annotate references to mythological entities in literature. The integration of LLMs into these tasks not only improves the efficiency and scalability of computational methods but also opens up new avenues for research by automating processes that traditionally required extensive manual effort. However, challenges remain, particularly in ensuring the accuracy and reliability of LLM outputs and addressing ethical concerns related to their use.

Noteworthy Papers

• PROPOE 2: Extends the capabilities of computational poetry generation by exploring a wider range of rhythms and sound effects, demonstrating the system's ability to produce varied and coherent poems.
• Automatic Extraction of Metaphoric Analogies from Literary Texts: Introduces a novel dataset and evaluates the performance of LLMs in extracting and generating elements of metaphoric analogies, showcasing their potential in automating complex literary analysis tasks.
• A Dual-Perspective Metaphor Detection Framework Using Large Language Models: Proposes a new framework that enhances the transparency and reliability of metaphor detection by LLMs, achieving state-of-the-art performance.
• Annotating References to Mythological Entities in French Literature: Explores the application of LLMs for annotating mythological references, highlighting both their capabilities and limitations in literary analysis.