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

ArcaneQA: Dynamic Program Induction and Contextualized Encoding for Knowledge Base Question Answering

About

Question answering on knowledge bases (KBQA) poses a unique challenge for semantic parsing research due to two intertwined challenges: large search space and ambiguities in schema linking. Conventional ranking-based KBQA models, which rely on a candidate enumeration step to reduce the search space, struggle with flexibility in predicting complicated queries and have impractical running time. In this paper, we present ArcaneQA, a novel generation-based model that addresses both the large search space and the schema linking challenges in a unified framework with two mutually boosting ingredients: dynamic program induction for tackling the large search space and dynamic contextualized encoding for schema linking. Experimental results on multiple popular KBQA datasets demonstrate the highly competitive performance of ArcaneQA in both effectiveness and efficiency.

Yu Gu, Yu Su• 2022

Related benchmarks

TaskDatasetResultRank
Knowledge Base Question AnsweringWEBQSP (test)--
143
Knowledge Base Question AnsweringWebQSP Freebase (test)
F1 Score75.6
46
Knowledge Base Question AnsweringGrailQA v1.0 (test)
Overall EM63.8
33
Knowledge Base Question AnsweringGrailQA (test)
F181.81
27
Knowledge Base Question AnsweringGraphQ (test)
F147.5
19
Knowledge Base Question AnsweringWebQSP v1 (test)
F1 Score75.6
16
Knowledge Base Question AnsweringGrailQA v1.0 (dev)
F176.9
9
Knowledge Base Question AnsweringGraphQ
F1 Score34.3
9
Showing 8 of 8 rows

Other info

Follow for update