SYGMA: A System for Generalizable and Modular Question Answering Over Knowledge Bases
Sumit Neelam, Udit Sharma, Hima Karanam, Shajith Ikbal, Pavan Kapanipathi, Ibrahim Abdelaziz, Nandana Mihindukulasooriya, Young-Suk Lee
Abstract
Knowledge Base Question Answering (KBQA) involving complex reasoning is emerging as an important research direction. However, most KBQA systems struggle with generalizability, particularly on two dimensions: (a) across multiple knowledge bases, where existing KBQA approaches are typically tuned to a single knowledge base, and (b) across multiple reasoning types, where majority of datasets and systems have primarily focused on multi-hop reasoning. In this paper, we present SYGMA, a modular KBQA approach developed with goal of generalization across multiple knowledge bases and multiple reasoning types. To facilitate this, SYGMA is designed as two high level modules: 1) KB-agnostic question understanding module that remain common across KBs, and generates logic representation of the question with high level reasoning constructs that are extensible, and 2) KB-specific question mapping and answering module to address the KB-specific aspects of the answer extraction. We evaluated SYGMA on multiple datasets belonging to distinct knowledge bases (DBpedia and Wikidata) and distinct reasoning types (multi-hop and temporal). State-of-the-art or competitive performances achieved on those datasets demonstrate its generalization capability.
BibTeX
@inproceedings{neelam-etal-2022-sygma,
title = "{SYGMA}: A System for Generalizable and Modular Question Answering Over Knowledge Bases",
author = "Neelam, Sumit and
Sharma, Udit and
Karanam, Hima and
Ikbal, Shajith and
Kapanipathi, Pavan and
Abdelaziz, Ibrahim and
Mihindukulasooriya, Nandana and
Lee, Young-Suk and
Srivastava, Santosh and
Pendus, Cezar and
Dana, Saswati and
Garg, Dinesh and
Fokoue, Achille and
Bhargav, G P Shrivatsa and
Khandelwal, Dinesh and
Ravishankar, Srinivas and
Gurajada, Sairam and
Chang, Maria and
Uceda-Sosa, Rosario and
Roukos, Salim and
Gray, Alexander and
Lima, Guilherme and
Riegel, Ryan and
Luus, Francois and
Subramaniam, L V",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.284/",
doi = "10.18653/v1/2022.findings-emnlp.284",
pages = "3866--3879"
}