Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval
Abstract
We propose a simple and efficient multi-hop dense retrieval approach for answering complex open-domain questions, which achieves state-of-the-art performance on two multi-hop datasets, HotpotQA and multi-evidence FEVER. Contrary to previous work, our method does not require access to any corpus-specific information, such as inter-document hyperlinks or human-annotated entity markers, and can be applied to any unstructured text corpus. Our system also yields a much better efficiency-accuracy trade-off, matching the best published accuracy on HotpotQA while being 10 times faster at inference time.
Cite
Text
Xiong et al. "Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval." International Conference on Learning Representations, 2021.Markdown
[Xiong et al. "Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval." International Conference on Learning Representations, 2021.](https://mlanthology.org/iclr/2021/xiong2021iclr-answering/)BibTeX
@inproceedings{xiong2021iclr-answering,
title = {{Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval}},
author = {Xiong, Wenhan and Li, Xiang and Iyer, Srini and Du, Jingfei and Lewis, Patrick and Wang, William Yang and Mehdad, Yashar and Yih, Scott and Riedel, Sebastian and Kiela, Douwe and Oguz, Barlas},
booktitle = {International Conference on Learning Representations},
year = {2021},
url = {https://mlanthology.org/iclr/2021/xiong2021iclr-answering/}
}