# Repartitioning of the ComplexWebQuestions Dataset

arXiv: https://arxiv.org/abs/1807.09623 · HTML: https://mlsanity.com/paper/1807.09623/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jul 2018.

## Built on by (most cited first, 3 of 3)

- [Hopper: Multi-hop Transformer for Spatiotemporal Reasoning](https://mlsanity.com/paper/2103.10574/) (2021)
- [Towards Transparent Interactive Semantic Parsing via Step-by-Step Correction](https://mlsanity.com/paper/2110.08345/) (2021)
- [BMGQ: A Bottom-up Method for Generating Complex Multi-hop Reasoning Questions from Semi-structured Data](https://mlsanity.com/paper/2510.24151/) (2025)

## Abstract (arXiv)

Recently, Talmor and Berant (2018) introduced ComplexWebQuestions - a dataset focused on answering complex questions by decomposing them into a sequence of simpler questions and extracting the answer from retrieved web snippets. In their work the authors used a pre-trained reading comprehension (RC) model (Salant and Berant, 2018) to extract the answer from the web snippets. In this short note we show that training a RC model directly on the training data of ComplexWebQuestions reveals a leakage from the training set to the test set that allows to obtain unreasonably high performance. As a solution, we construct a new partitioning of ComplexWebQuestions that does not suffer from this leakage and publicly release it. We also perform an empirical evaluation on these two datasets and show that training a RC model on the training data substantially improves state-of-the-art performance.