We study open-domain question answering with structured, unstructured and semi-structured knowledge sources, including text, tables, lists and knowledge bases. Departing from prior work, we propose a unifying approach that homogenizes all sources by reducing them to text and applies the retriever-reader model which has so far been limited to text sources only. Our approach greatly improves the results on knowledge-base QA tasks by 11 points, compared to latest graph-based methods. More importantly, we demonstrate that our unified knowledge (UniK-QA) model is a simple and yet effective way to combine heterogeneous sources of knowledge, advancing the state-of-the-art results on two popular question answering benchmarks, NaturalQuestions and WebQuestions, by 3.5 and 2.6 points, respectively. The code of UniK-QA is available at:https://github.com/facebookresearch/UniK-QA.
@inproceedings{oguz-etal-2022-unik, title = "{U}ni{K}-{QA}: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering", author = "Oguz, Barlas and Chen, Xilun and Karpukhin, Vladimir and Peshterliev, Stan and Okhonko, Dmytro and Schlichtkrull, Michael and Gupta, Sonal and Mehdad, Yashar and Yih, Scott", editor = "Carpuat, Marine and de Marneffe, Marie-Catherine and Meza Ruiz, Ivan Vladimir", booktitle = "Findings of the Association for Computational Linguistics: NAACL 2022", month = jul, year = "2022", address = "Seattle, United States", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2022.findings-naacl.115/", doi = "10.18653/v1/2022.findings-naacl.115", pages = "1535--1546", abstract = "We study open-domain question answering with structured, unstructured and semi-structured knowledge sources, including text, tables, lists and knowledge bases. Departing from prior work, we propose a unifying approach that homogenizes all sources by reducing them to text and applies the retriever-reader model which has so far been limited to text sources only. Our approach greatly improves the results on knowledge-base QA tasks by 11 points, compared to latest graph-based methods. More importantly, we demonstrate that our unified knowledge (UniK-QA) model is a simple and yet effective way to combine heterogeneous sources of knowledge, advancing the state-of-the-art results on two popular question answering benchmarks, NaturalQuestions and WebQuestions, by 3.5 and 2.6 points, respectively. The code of UniK-QA is available at: \url{https://github.com/facebookresearch/UniK-QA}."}
%0 Conference Proceedings%T UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering%A Oguz, Barlas%A Chen, Xilun%A Karpukhin, Vladimir%A Peshterliev, Stan%A Okhonko, Dmytro%A Schlichtkrull, Michael%A Gupta, Sonal%A Mehdad, Yashar%A Yih, Scott%Y Carpuat, Marine%Y de Marneffe, Marie-Catherine%Y Meza Ruiz, Ivan Vladimir%S Findings of the Association for Computational Linguistics: NAACL 2022%D 2022%8 July%I Association for Computational Linguistics%C Seattle, United States%F oguz-etal-2022-unik%X We study open-domain question answering with structured, unstructured and semi-structured knowledge sources, including text, tables, lists and knowledge bases. Departing from prior work, we propose a unifying approach that homogenizes all sources by reducing them to text and applies the retriever-reader model which has so far been limited to text sources only. Our approach greatly improves the results on knowledge-base QA tasks by 11 points, compared to latest graph-based methods. More importantly, we demonstrate that our unified knowledge (UniK-QA) model is a simple and yet effective way to combine heterogeneous sources of knowledge, advancing the state-of-the-art results on two popular question answering benchmarks, NaturalQuestions and WebQuestions, by 3.5 and 2.6 points, respectively. The code of UniK-QA is available at: https://github.com/facebookresearch/UniK-QA.%R 10.18653/v1/2022.findings-naacl.115%U https://aclanthology.org/2022.findings-naacl.115/%U https://doi.org/10.18653/v1/2022.findings-naacl.115%P 1535-1546
[UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering](https://aclanthology.org/2022.findings-naacl.115/) (Oguz et al., Findings 2022)