DeDisCo at the DISRPT 2025 Shared Task: A System for Discourse Relation Classification
This is an incremental improvement for the specific NLP community focused on discourse analysis.
The paper tackled discourse relation classification in the DISRPT 2025 shared task by testing two approaches with mt5-based and Qwen models, plus data augmentation and linguistic features, achieving a macro-accuracy score of 71.28.
This paper presents DeDisCo, Georgetown University's entry in the DISRPT 2025 shared task on discourse relation classification. We test two approaches, using an mt5-based encoder and a decoder based approach using the openly available Qwen model. We also experiment on training with augmented dataset for low-resource languages using matched data translated automatically from English, as well as using some additional linguistic features inspired by entries in previous editions of the Shared Task. Our system achieves a macro-accuracy score of 71.28, and we provide some interpretation and error analysis for our results.