SF-QA: Simple and Fair Evaluation Library for Open-domain Question Answering

TL;DR

SF-QA is a modular, accessible, and reproducible evaluation framework for open-domain question answering, designed to lower resource barriers and facilitate community contributions.

Contribution

It introduces a modular evaluation library that simplifies and standardizes open-domain QA system assessment, promoting fairness and reproducibility.

Findings

Framework is publicly available and open for contributions

Modular design enhances accessibility and reproducibility

Facilitates fair comparison of open-domain QA systems

Abstract

Although open-domain question answering (QA) draws great attention in recent years, it requires large amounts of resources for building the full system and is often difficult to reproduce previous results due to complex configurations. In this paper, we introduce SF-QA: simple and fair evaluation framework for open-domain QA. SF-QA framework modularizes the pipeline open-domain QA system, which makes the task itself easily accessible and reproducible to research groups without enough computing resources. The proposed evaluation framework is publicly available and anyone can contribute to the code and evaluations.