Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answering

Answering questions that require multi-hop reasoning at web-scale necessitates retrieving multiple evidence documents, one of which often has little lexical or semantic relationship to the question. This paper introduces a new graph-based recurrent retrieval approach that learns to retrieve reasoning paths over the Wikipedia graph to answer multi-hop open-domain questions. Our retriever model trains a recurrent neural network that learns to sequentially retrieve evidence paragraphs in the reasoning path by conditioning on the previously retrieved documents. Our reader model ranks the reasoning paths and extracts the answer span included in the best reasoning path. Experimental results show state-of-the-art results in three open-domain QA datasets, showcasing the effectiveness and robustness of our method. Notably, our method achieves significant improvement in HotpotQA, outperforming the previous best model by more than 14 points.

PDF Abstract ICLR 2020 PDF ICLR 2020 Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Question Answering HotpotQA Robustly Fine-tuned Graph-based Recurrent Retriever ANS-EM 0.600 # 25
ANS-F1 0.730 # 25
SUP-EM 0.491 # 27
SUP-F1 0.764 # 28
JOINT-EM 0.354 # 26
JOINT-F1 0.612 # 27

Methods


No methods listed for this paper. Add relevant methods here