KG^2: Learning to Reason Science Exam Questions with Contextual Knowledge Graph Embeddings
This addresses the problem of improving AI reasoning in science question answering, representing an incremental advance over existing methods.
The paper tackles the challenge of answering difficult science exam questions from the ARC benchmark by proposing a framework that mimics human open-book problem-solving, using contextual knowledge graphs for questions and supporting sentences. Their model outperforms previous state-of-the-art QA systems on the ARC Challenge Set.
The AI2 Reasoning Challenge (ARC), a new benchmark dataset for question answering (QA) has been recently released. ARC only contains natural science questions authored for human exams, which are hard to answer and require advanced logic reasoning. On the ARC Challenge Set, existing state-of-the-art QA systems fail to significantly outperform random baseline, reflecting the difficult nature of this task. In this paper, we propose a novel framework for answering science exam questions, which mimics human solving process in an open-book exam. To address the reasoning challenge, we construct contextual knowledge graphs respectively for the question itself and supporting sentences. Our model learns to reason with neural embeddings of both knowledge graphs. Experiments on the ARC Challenge Set show that our model outperforms the previous state-of-the-art QA systems.