A Logic-Driven Framework for Consistency of Neural Models

Tao Li, Vivek Gupta, Maitrey Mehta, Vivek Srikumar


Abstract
While neural models show remarkable accuracy on individual predictions, their internal beliefs can be inconsistent across examples. In this paper, we formalize such inconsistency as a generalization of prediction error. We propose a learning framework for constraining models using logic rules to regularize them away from inconsistency. Our framework can leverage both labeled and unlabeled examples and is directly compatible with off-the-shelf learning schemes without model redesign. We instantiate our framework on natural language inference, where experiments show that enforcing invariants stated in logic can help make the predictions of neural models both accurate and consistent.
Anthology ID:
D19-1405
Volume:
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
Month:
November
Year:
2019
Address:
Hong Kong, China
Venues:
EMNLP | IJCNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
3924–3935
Language:
URL:
https://aclanthology.org/D19-1405
DOI:
10.18653/v1/D19-1405
Bibkey:
Cite (ACL):
Tao Li, Vivek Gupta, Maitrey Mehta, and Vivek Srikumar. 2019. A Logic-Driven Framework for Consistency of Neural Models. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3924–3935, Hong Kong, China. Association for Computational Linguistics.
Cite (Informal):
A Logic-Driven Framework for Consistency of Neural Models (Li et al., EMNLP 2019)
Copy Citation:
PDF:
https://aclanthology.org/D19-1405.pdf
Code
 utahnlp/consistency
Data
GLUEMultiNLISNLI