Complex Reading Comprehension Through Question Decomposition
This addresses the problem of complex reasoning in reading comprehension for NLP researchers, though it is incremental as it builds on existing methods.
The paper tackles multi-hop reading comprehension by proposing a learning approach that decomposes complex questions into sub-questions, improving performance with a 7.2 absolute F1 point gain on a hard subset of the DROP dataset.
Multi-hop reading comprehension requires not only the ability to reason over raw text but also the ability to combine multiple evidence. We propose a novel learning approach that helps language models better understand difficult multi-hop questions and perform "complex, compositional" reasoning. Our model first learns to decompose each multi-hop question into several sub-questions by a trainable question decomposer. Instead of answering these sub-questions, we directly concatenate them with the original question and context, and leverage a reading comprehension model to predict the answer in a sequence-to-sequence manner. By using the same language model for these two components, our best seperate/unified t5-base variants outperform the baseline by 7.2/6.1 absolute F1 points on a hard subset of DROP dataset.