arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2102.08345cs.CL
收藏

NoiseQA:面向以用户为中心的问答的挑战集评估

NoiseQA: Challenge Set Evaluation for User-Centric Question Answering

Carnegie Mellon University(卡内基梅隆大学)

Abhilasha Ravichander, Siddharth Dalmia, Maria Ryskina, Florian Metze, Eduard Hovy, Alan W Black

摘要:

当在现实世界中部署问答(QA)系统时,用户通过各种接口来查询它们,例如与语音助理交谈,将问题键入搜索引擎,甚至将问题翻译为QA系统支持的语言。虽然有大量的社区关注致力于在假设完美形成的问题的段落中识别正确答案,但我们表明,在回答引擎之前的管道中的组件可能会引入各种各样的大量错误来源,并且性能可能会根据这些上游噪声源大幅下降,即使是强大的预训练QA模型。我们的结论是,在QA系统可以有效地部署之前,还有很大的进步空间,强调QA评估需要扩展到考虑现实世界的使用,并希望我们的研究结果将激发社区对我们的系统实际上需要对人类有用时出现的问题的更大兴趣。

英文摘要:

When Question-Answering (QA) systems are deployed in the real world, users query them through a variety of interfaces, such as speaking to voice assistants, typing questions into a search engine, or even translating questions to languages supported by the QA system. While there has been significant community attention devoted to identifying correct answers in passages assuming a perfectly formed question, we show that components in the pipeline that precede an answering engine can introduce varied and considerable sources of error, and performance can degrade substantially based on these upstream noise sources even for powerful pre-trained QA models. We conclude that there is substantial room for progress before QA systems can be effectively deployed, highlight the need for QA evaluation to expand to consider real-world use, and hope that our findings will spur greater community interest in the issues that arise when our systems actually need to be of utility to humans.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑