DAGN: Discourse-Aware Graph Network for Logical Reasoning
This work addresses logical reasoning QA for NLP researchers, but it is incremental as it builds on existing graph-based methods with discourse-aware features.
The paper tackles the problem of logical reasoning in question answering by incorporating passage-level discourse relations, proposing DAGN which encodes discourse structure as a graph and learns features for QA tasks. It achieves competitive results on ReClor and LogiQA datasets, though specific numerical gains are not provided.
Recent QA with logical reasoning questions requires passage-level relations among the sentences. However, current approaches still focus on sentence-level relations interacting among tokens. In this work, we explore aggregating passage-level clues for solving logical reasoning QA by using discourse-based information. We propose a discourse-aware graph network (DAGN) that reasons relying on the discourse structure of the texts. The model encodes discourse information as a graph with elementary discourse units (EDUs) and discourse relations, and learns the discourse-aware features via a graph network for downstream QA tasks. Experiments are conducted on two logical reasoning QA datasets, ReClor and LogiQA, and our proposed DAGN achieves competitive results. The source code is available at https://github.com/Eleanor-H/DAGN.