Conversational Multi-Hop Reasoning with Neural Commonsense Knowledge and Symbolic Logic Rules

Forough Arabshahi, Jennifer Lee, Antoine Bosselut, Yejin Choi, Tom Mitchell


Abstract
One of the challenges faced by conversational agents is their inability to identify unstated presumptions of their users’ commands, a task trivial for humans due to their common sense. In this paper, we propose a zero-shot commonsense reasoning system for conversational agents in an attempt to achieve this. Our reasoner uncovers unstated presumptions from user commands satisfying a general template of if-(state), then-(action), because-(goal). Our reasoner uses a state-of-the-art transformer-based generative commonsense knowledge base (KB) as its source of background knowledge for reasoning. We propose a novel and iterative knowledge query mechanism to extract multi-hop reasoning chains from the neural KB which uses symbolic logic rules to significantly reduce the search space. Similar to any KBs gathered to date, our commonsense KB is prone to missing knowledge. Therefore, we propose to conversationally elicit the missing knowledge from human users with our novel dynamic question generation strategy, which generates and presents contextualized queries to human users. We evaluate the model with a user study with human users that achieves a 35% higher success rate compared to SOTA.
Anthology ID:
2021.emnlp-main.588
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7404–7418
Language:
URL:
https://aclanthology.org/2021.emnlp-main.588
DOI:
10.18653/v1/2021.emnlp-main.588
Bibkey:
Cite (ACL):
Forough Arabshahi, Jennifer Lee, Antoine Bosselut, Yejin Choi, and Tom Mitchell. 2021. Conversational Multi-Hop Reasoning with Neural Commonsense Knowledge and Symbolic Logic Rules. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 7404–7418, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Conversational Multi-Hop Reasoning with Neural Commonsense Knowledge and Symbolic Logic Rules (Arabshahi et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.588.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.588.mp4
Data
ConceptNet