Differentiable Open-Ended Commonsense Reasoning

Bill Yuchen Lin, Haitian Sun, Bhuwan Dhingra, Manzil Zaheer, Xiang Ren, William Cohen


Abstract
Current commonsense reasoning research focuses on developing models that use commonsense knowledge to answer multiple-choice questions. However, systems designed to answer multiple-choice questions may not be useful in applications that do not provide a small list of candidate answers to choose from. As a step towards making commonsense reasoning research more realistic, we propose to study open-ended commonsense reasoning (OpenCSR) — the task of answering a commonsense question without any pre-defined choices — using as a resource only a corpus of commonsense facts written in natural language. OpenCSR is challenging due to a large decision space, and because many questions require implicit multi-hop reasoning. As an approach to OpenCSR, we propose DrFact, an efficient Differentiable model for multi-hop Reasoning over knowledge Facts. To evaluate OpenCSR methods, we adapt several popular commonsense reasoning benchmarks, and collect multiple new answers for each test question via crowd-sourcing. Experiments show that DrFact outperforms strong baseline methods by a large margin.
Anthology ID:
2021.naacl-main.366
Volume:
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Month:
June
Year:
2021
Address:
Online
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4611–4625
Language:
URL:
https://aclanthology.org/2021.naacl-main.366
DOI:
10.18653/v1/2021.naacl-main.366
Bibkey:
Cite (ACL):
Bill Yuchen Lin, Haitian Sun, Bhuwan Dhingra, Manzil Zaheer, Xiang Ren, and William Cohen. 2021. Differentiable Open-Ended Commonsense Reasoning. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 4611–4625, Online. Association for Computational Linguistics.
Cite (Informal):
Differentiable Open-Ended Commonsense Reasoning (Lin et al., NAACL 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.naacl-main.366.pdf
Video:
 https://preview.aclanthology.org/ingestion-script-update/2021.naacl-main.366.mp4
Data
ARCGenericsKBHotpotQANatural QuestionsQASC