Winnowing Knowledge for Multi-choice Question Answering

Yeqiu Li, Bowei Zou, Zhifeng Li, Ai Ti Aw, Yu Hong, Qiaoming Zhu


Abstract
We tackle multi-choice question answering. Acquiring related commonsense knowledge to the question and options facilitates the recognition of the correct answer. However, the current reasoning models suffer from the noises in the retrieved knowledge. In this paper, we propose a novel encoding method which is able to conduct interception and soft filtering. This contributes to the harvesting and absorption of representative information with less interference from noises. We experiment on CommonsenseQA. Experimental results illustrate that our method yields substantial and consistent improvements compared to the strong Bert, RoBERTa and Albert-based baselines.
Anthology ID:
2021.findings-emnlp.100
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2021
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Venue:
Findings
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
1157–1165
Language:
URL:
https://aclanthology.org/2021.findings-emnlp.100
DOI:
10.18653/v1/2021.findings-emnlp.100
Bibkey:
Cite (ACL):
Yeqiu Li, Bowei Zou, Zhifeng Li, Ai Ti Aw, Yu Hong, and Qiaoming Zhu. 2021. Winnowing Knowledge for Multi-choice Question Answering. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 1157–1165, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Winnowing Knowledge for Multi-choice Question Answering (Li et al., Findings 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.findings-emnlp.100.pdf
Software:
 2021.findings-emnlp.100.Software.zip
Video:
 https://preview.aclanthology.org/ingestion-script-update/2021.findings-emnlp.100.mp4
Data
CommonsenseQAConceptNet