Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference

Yufei Feng, Xiaoyu Yang, Xiaodan Zhu, Michael Greenspan


Abstract
We introduce a neuro-symbolic natural logic framework based on reinforcement learning with introspective revision. The model samples and rewards specific reasoning paths through policy gradient, in which the introspective revision algorithm modifies intermediate symbolic reasoning steps to discover reward-earning operations as well as leverages external knowledge to alleviate spurious reasoning and training inefficiency. The framework is supported by properly designed local relation models to avoid input entangling, which helps ensure the interpretability of the proof paths. The proposed model has built-in interpretability and shows superior capability in monotonicity inference, systematic generalization, and interpretability, compared with previous models on the existing datasets.
Anthology ID:
2022.tacl-1.14
Volume:
Transactions of the Association for Computational Linguistics, Volume 10
Month:
Year:
2022
Address:
Cambridge, MA
Editors:
Brian Roark, Ani Nenkova
Venue:
TACL
SIG:
Publisher:
MIT Press
Note:
Pages:
240–256
Language:
URL:
https://aclanthology.org/2022.tacl-1.14
DOI:
10.1162/tacl_a_00458
Bibkey:
Cite (ACL):
Yufei Feng, Xiaoyu Yang, Xiaodan Zhu, and Michael Greenspan. 2022. Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference. Transactions of the Association for Computational Linguistics, 10:240–256.
Cite (Informal):
Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference (Feng et al., TACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.tacl-1.14.pdf