Everything Happens for a Reason: Discovering the Purpose of Actions in Procedural Text
This work addresses the challenge of understanding procedural text for AI systems, such as in education or automation, but it is incremental as it builds on an existing framework and dataset.
The paper tackles the problem of comprehending procedural text by predicting not only what actions occur but also why they happen in a certain order, introducing a new model (XPAD) that biases effect predictions to explain more actions and align with background knowledge. The result is that XPAD significantly outperforms prior systems on the new task of explaining action dependencies while maintaining performance on the original task, with the dataset extended and made available.
Our goal is to better comprehend procedural text, e.g., a paragraph about photosynthesis, by not only predicting what happens, but why some actions need to happen before others. Our approach builds on a prior process comprehension framework for predicting actions' effects, to also identify subsequent steps that those effects enable. We present our new model (XPAD) that biases effect predictions towards those that (1) explain more of the actions in the paragraph and (2) are more plausible with respect to background knowledge. We also extend an existing benchmark dataset for procedural text comprehension, ProPara, by adding the new task of explaining actions by predicting their dependencies. We find that XPAD significantly outperforms prior systems on this task, while maintaining the performance on the original task in ProPara. The dataset is available at http://data.allenai.org/propara