Discourse Parsing

Discourse parsing aims to uncover the hierarchical structure of text, revealing relationships between sentences or clauses beyond simple sentence-level analysis. Current research focuses on improving accuracy and efficiency through various approaches, including joint training frameworks that consider multiple theoretical criteria simultaneously, the application of large language models (LLMs) for both top-down and bottom-up parsing strategies, and the development of unsupervised and distant supervision methods to address data scarcity. These advancements are crucial for enhancing numerous natural language processing applications, such as dialogue systems, argument mining, and machine translation, by enabling a deeper understanding of textual coherence and meaning.