Red Dragon AI at TextGraphs 2019 Shared Task: Language Model Assisted Explanation Generation

TL;DR

This paper describes Red Dragon AI's participation in the TextGraphs 2019 Shared Task, focusing on generating explanations for elementary science questions using language models, with three progressively sophisticated methods.

Contribution

The paper introduces three novel methods for explanation generation that leverage language models directly, improving performance over previous approaches.

Findings

Placed 3rd in the competition leaderboard.

Developed three methods with increasing sophistication.

Achieved higher scores with each successive method.

Abstract

The TextGraphs-13 Shared Task on Explanation Regeneration asked participants to develop methods to reconstruct gold explanations for elementary science questions. Red Dragon AI's entries used the language of the questions and explanation text directly, rather than a constructing a separate graph-like representation. Our leaderboard submission placed us 3rd in the competition, but we present here three methods of increasing sophistication, each of which scored successively higher on the test set after the competition close.