arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.05883cs.CL
收藏

基于Beam稠密检索的非结构化文本多步推理

Multi-Step Reasoning Over Unstructured Text with Beam Dense Retrieval

University of Maryland(马里兰大学);Microsoft Research(微软研究院)

Chen Zhao, Chenyan Xiong, Jordan Boyd-Graber, Hal Daumé

摘要:

复杂的问答通常需要找到由多个证据组成的推理链。目前的方法结合了结构化知识和非结构化文本的优势,假设文本语料库是半结构化的。基于稠密检索方法,提出了一种新的多步检索方法(BeamDR),通过密集表示中的束搜索迭代形成证据链。当在多跳问答上进行评估时,BeamDR与最先进的系统相比具有竞争力,而不使用任何半结构化信息。通过密集空间中的查询组合,BeamDR捕获推理链中证据之间的隐含关系。该代码可在https://github.com/ henryzhao 5852/BeamDR上获得。

英文摘要:

Complex question answering often requires finding a reasoning chain that consists of multiple evidence pieces. Current approaches incorporate the strengths of structured knowledge and unstructured text, assuming text corpora is semi-structured. Building on dense retrieval methods, we propose a new multi-step retrieval approach (BeamDR) that iteratively forms an evidence chain through beam search in dense representations. When evaluated on multi-hop question answering, BeamDR is competitive to state-of-the-art systems, without using any semi-structured information. Through query composition in dense space, BeamDR captures the implicit relationships between evidence in the reasoning chain. The code is available at https://github.com/ henryzhao5852/BeamDR.

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。

↑