arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2106.01074cs.CLcs.AIcs.DB
收藏

文本数据库推理

Database Reasoning Over Text

University of Cambridge(剑桥大学);Facebook AI(脸书人工智能);Sapienza University, Rome(罗马萨皮恩扎大学);University College London(伦敦大学学院)

James Thorne, Majid Yazdani, Marzieh Saeidi, Fabrizio Silvestri, Sebastian Riedel, Alon Halevy

摘要:

神经模型在回答来自自然语言文本的查询方面表现出令人印象深刻的性能提升。然而,现有的作品无法支持数据库查询,如“列出/计数所有女运动员谁出生在20世纪”,这需要推理的相关事实集的操作,如加入,过滤和聚合。我们表明,虽然国家的最先进的Transformer模型表现非常好的小型数据库,他们表现出的局限性,在处理嘈杂的数据,数值运算,查询聚合的事实。我们提出了一个模块化的体系结构,以回答这些数据库风格的查询多个跨度从文本和聚合这些规模。我们使用WikiNLDB,一个新的数据集探索这样的查询的架构进行评估。我们的架构可以扩展到包含数千个事实的数据库,而当代模型则受到可以编码多少事实的限制。在小型数据库的直接比较中,我们的方法将整体答案准确率从85%提高到90%。在更大的数据库上,我们的方法保持了它的准确性,而Transformer基线不能对上下文进行编码。

英文摘要:

Neural models have shown impressive performance gains in answering queries from natural language text. However, existing works are unable to support database queries, such as "List/Count all female athletes who were born in 20th century", which require reasoning over sets of relevant facts with operations such as join, filtering and aggregation. We show that while state-of-the-art transformer models perform very well for small databases, they exhibit limitations in processing noisy data, numerical operations, and queries that aggregate facts. We propose a modular architecture to answer these database-style queries over multiple spans from text and aggregating these at scale. We evaluate the architecture using WikiNLDB, a novel dataset for exploring such queries. Our architecture scales to databases containing thousands of facts whereas contemporary models are limited by how many facts can be encoded. In direct comparison on small databases, our approach increases overall answer accuracy from 85% to 90%. On larger databases, our approach retains its accuracy whereas transformer baselines could not encode the context.

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。

↑