Modeling Content and Context with Deep Relational Learning

TL;DR

This paper introduces DRaiL, a flexible framework that combines neural and symbolic methods for deep relational modeling in NLP, enabling better handling of complex language tasks.

Contribution

The paper presents DRaiL, a novel declarative framework that supports deep relational models tailored for diverse NLP applications, integrating language encoders and reasoning.

Findings

Supports integration with expressive language encoders

Facilitates study of interactions between representation, inference, and learning

Open-source implementation available

Abstract

Building models for realistic natural language tasks requires dealing with long texts and accounting for complicated structural dependencies. Neural-symbolic representations have emerged as a way to combine the reasoning capabilities of symbolic methods, with the expressiveness of neural networks. However, most of the existing frameworks for combining neural and symbolic representations have been designed for classic relational learning tasks that work over a universe of symbolic entities and relations. In this paper, we present DRaiL, an open-source declarative framework for specifying deep relational models, designed to support a variety of NLP scenarios. Our framework supports easy integration with expressive language encoders, and provides an interface to study the interactions between representation, inference and learning.