Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

UniK-QA: Unified Representations of Structured and Unstructured Knowledge for Open-Domain Question Answering

About

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.

Barlas Oguz, Xilun Chen, Vladimir Karpukhin, Stan Peshterliev, Dmytro Okhonko, Michael Schlichtkrull, Sonal Gupta, Yashar Mehdad, Scott Yih• 2020

Related benchmarks

TaskDatasetResultRank
Knowledge Base Question AnsweringWEBQSP (test)
Hit@179.1
143
Open Question AnsweringNatural Questions (NQ) (test)
Exact Match (EM)54.6
134
Open-domain Question AnsweringTriviaQA (test)
Exact Match65.1
80
Open-domain Question AnsweringWebQuestions (WebQ) (test)
Exact Match (EM)57.8
55
End-to-end Open-Domain Question AnsweringNQ (test)
Exact Match (EM)54
50
Knowledge Base Question AnsweringWebQSP Freebase (test)
F1 Score79.1
46
Open-domain Question AnsweringCuratedTREC (test)
Exact Match (EM)55.3
26
Temporal Question AnsweringTIME QUESTIONS 1.0 (test)
P@142.4
18
Temporal Question AnsweringTIQ 1.0 (test)
P@10.425
10
Open-domain Question AnsweringCOMPMIX (test)
Exact Match44
9
Showing 10 of 13 rows

Other info

Code

Follow for update