# ANTIQUE: A Non-Factoid Question Answering Benchmark

## TL;DR

ANTIQUE is a large-scale, real-user non-factoid question answering dataset with extensive relevance annotations, designed to advance research in answer passage retrieval for modern information retrieval systems.

## Contribution

The paper introduces ANTIQUE, a new large-scale non-factoid QA dataset with real questions and relevance judgments, filling a critical gap in existing resources.

## Key findings

- Baseline models show significant room for improvement.
- Neural IR models outperform classical approaches.
- Data analysis reveals diverse question categories.

## Abstract

Considering the widespread use of mobile and voice search, answer passage retrieval for non-factoid questions plays a critical role in modern information retrieval systems. Despite the importance of the task, the community still feels the significant lack of large-scale non-factoid question answering collections with real questions and comprehensive relevance judgments. In this paper, we develop and release a collection of 2,626 open-domain non-factoid questions from a diverse set of categories. The dataset, called ANTIQUE, contains 34,011 manual relevance annotations. The questions were asked by real users in a community question answering service, i.e., Yahoo! Answers. Relevance judgments for all the answers to each question were collected through crowdsourcing. To facilitate further research, we also include a brief analysis of the data as well as baseline results on both classical and recently developed neural IR models.