Bi-directional Attention Flow (BiDAF) network is a multi-stage hierarchical process that represents context at different levels of granularity and uses a bi-directional attention flow mechanism to achieve a query-aware context representation without early summarization.
This work explores the use of Markov Logic Networks (MLNs) for answering
elementary-level natural language science questions. The dataset contains the MLNs
generated from three different formulations along with a README describing the format.
Also see "Markov Logic Networks for Natural Language Question Answering" (StarAI '15) for a description of the formulations.