Discourse Relation

Discourse relation research focuses on understanding how sentences and clauses connect to form coherent text, aiming to automatically identify these relationships (explicit or implicit) and their types. Current research emphasizes improving the accuracy and efficiency of models, often employing multi-label classification, joint training frameworks, and contrastive learning methods, sometimes incorporating hierarchical sense structures and leveraging pre-trained language models. This work is significant for advancing natural language processing, enabling improved text summarization, question answering, and machine translation, as well as offering insights into human language comprehension and cognitive processes.