Abstract
We recently introduced DRaiL, a declarative neuro-symbolic modeling framework designed to support a wide variety of NLP scenarios. In this demo, we enhance DRaiL with an easy to use Python interface equipped with methods to define, modify and augment models interactively, as well as with methods to debug and visualize the predictions made. We demonstrate this interface with two challenging NLP tasks: analyzing moral sentiment in political discourse, and analyzing opinions about the Covid-19 vaccine.
Cite
CITATION STYLE
Pacheco, M. L., Roy, S., & Goldwasser, D. (2022). Hands-On Interactive Neuro-Symbolic NLP with DRaiL. In EMNLP 2022 - 2022 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Demonstrations Session (pp. 371–378). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2022.emnlp-demos.37
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
