# ReQA: An Evaluation for End-to-End Answer Retrieval Models

## TL;DR

ReQA introduces a new benchmark for evaluating end-to-end answer retrieval models at scale, emphasizing the importance of scalable, sentence-level retrieval methods beyond traditional passage-based QA.

## Contribution

The paper presents ReQA, a novel benchmark for large-scale answer retrieval, and provides baseline evaluations using neural and classical retrieval techniques.

## Key findings

- Baseline neural models show competitive performance.
- Classical retrieval methods remain effective for large-scale retrieval.
- ReQA facilitates future research in scalable answer retrieval.

## Abstract

Popular QA benchmarks like SQuAD have driven progress on the task of identifying answer spans within a specific passage, with models now surpassing human performance. However, retrieving relevant answers from a huge corpus of documents is still a challenging problem, and places different requirements on the model architecture. There is growing interest in developing scalable answer retrieval models trained end-to-end, bypassing the typical document retrieval step. In this paper, we introduce Retrieval Question-Answering (ReQA), a benchmark for evaluating large-scale sentence-level answer retrieval models. We establish baselines using both neural encoding models as well as classical information retrieval techniques. We release our evaluation code to encourage further work on this challenging task.