Reasoning Over Virtual Knowledge Bases With Open Predicate Relations

Haitian Sun, Patrick Verga, Bhuwan Dhingra, Ruslan Salakhutdinov, William W Cohen
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:9966-9977, 2021.

Abstract

We present the Open Predicate Query Language (OPQL); a method for constructing a virtual KB (VKB) trained entirely from text. Large Knowledge Bases (KBs) are indispensable for a wide-range of industry applications such as question answering and recommendation. Typically, KBs encode world knowledge in a structured, readily accessible form derived from laborious human annotation efforts. Unfortunately, while they are extremely high precision, KBs are inevitably highly incomplete and automated methods for enriching them are far too inaccurate. Instead, OPQL constructs a VKB by encoding and indexing a set of relation mentions in a way that naturally enables reasoning and can be trained without any structured supervision. We demonstrate that OPQL outperforms prior VKB methods on two different KB reasoning tasks and, additionally, can be used as an external memory integrated into a language model (OPQL-LM) leading to improvements on two open-domain question answering tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-sun21e, title = {Reasoning Over Virtual Knowledge Bases With Open Predicate Relations}, author = {Sun, Haitian and Verga, Patrick and Dhingra, Bhuwan and Salakhutdinov, Ruslan and Cohen, William W}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {9966--9977}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/sun21e/sun21e.pdf}, url = {https://proceedings.mlr.press/v139/sun21e.html}, abstract = {We present the Open Predicate Query Language (OPQL); a method for constructing a virtual KB (VKB) trained entirely from text. Large Knowledge Bases (KBs) are indispensable for a wide-range of industry applications such as question answering and recommendation. Typically, KBs encode world knowledge in a structured, readily accessible form derived from laborious human annotation efforts. Unfortunately, while they are extremely high precision, KBs are inevitably highly incomplete and automated methods for enriching them are far too inaccurate. Instead, OPQL constructs a VKB by encoding and indexing a set of relation mentions in a way that naturally enables reasoning and can be trained without any structured supervision. We demonstrate that OPQL outperforms prior VKB methods on two different KB reasoning tasks and, additionally, can be used as an external memory integrated into a language model (OPQL-LM) leading to improvements on two open-domain question answering tasks.} }
Endnote
%0 Conference Paper %T Reasoning Over Virtual Knowledge Bases With Open Predicate Relations %A Haitian Sun %A Patrick Verga %A Bhuwan Dhingra %A Ruslan Salakhutdinov %A William W Cohen %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-sun21e %I PMLR %P 9966--9977 %U https://proceedings.mlr.press/v139/sun21e.html %V 139 %X We present the Open Predicate Query Language (OPQL); a method for constructing a virtual KB (VKB) trained entirely from text. Large Knowledge Bases (KBs) are indispensable for a wide-range of industry applications such as question answering and recommendation. Typically, KBs encode world knowledge in a structured, readily accessible form derived from laborious human annotation efforts. Unfortunately, while they are extremely high precision, KBs are inevitably highly incomplete and automated methods for enriching them are far too inaccurate. Instead, OPQL constructs a VKB by encoding and indexing a set of relation mentions in a way that naturally enables reasoning and can be trained without any structured supervision. We demonstrate that OPQL outperforms prior VKB methods on two different KB reasoning tasks and, additionally, can be used as an external memory integrated into a language model (OPQL-LM) leading to improvements on two open-domain question answering tasks.
APA
Sun, H., Verga, P., Dhingra, B., Salakhutdinov, R. & Cohen, W.W.. (2021). Reasoning Over Virtual Knowledge Bases With Open Predicate Relations. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:9966-9977 Available from https://proceedings.mlr.press/v139/sun21e.html.

Related Material