English

Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts

Computation and Language 2022-11-01 v1

Abstract

Explicit decomposition modeling, which involves breaking down complex tasks into more straightforward and often more interpretable sub-tasks, has long been a central theme in developing robust and interpretable NLU systems. However, despite the many datasets and resources built as part of this effort, the majority have small-scale annotations and limited scope, which is insufficient to solve general decomposition tasks. In this paper, we look at large-scale intermediate pre-training of decomposition-based transformers using distant supervision from comparable texts, particularly large-scale parallel news. We show that with such intermediate pre-training, developing robust decomposition-based models for a diverse range of tasks becomes more feasible. For example, on semantic parsing, our model, DecompT5, improves 20% to 30% on two datasets, Overnight and TORQUE, over the baseline language model. We further use DecompT5 to build a novel decomposition-based QA system named DecompEntail, improving over state-of-the-art models, including GPT-3, on both HotpotQA and StrategyQA by 8% and 4%, respectively.

Keywords

Cite

@article{arxiv.2210.16865,
  title  = {Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Texts},
  author = {Ben Zhou and Kyle Richardson and Xiaodong Yu and Dan Roth},
  journal= {arXiv preprint arXiv:2210.16865},
  year   = {2022}
}

Comments

Accepted at EMNLP 2022

R2 v1 2026-06-28T04:47:49.106Z