Published on Tue Sep 10 2019

Everything Happens for a Reason: Discovering the Purpose of Actions in Procedural Text

Bhavana Dalvi Mishra, Niket Tandon, Antoine Bosselut, Wen-tau Yih, Peter Clark

Our goal is to better comprehend procedural text 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)

0
0
0
Abstract

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

Wed Aug 29 2018
Artificial Intelligence
Reasoning about Actions and State Changes by Injecting Commonsense Knowledge
Comprehending procedural text, e.g., a paragraph describing photosynthesis, requires modeling actions and the state changes they produce. Although several recent systems have shown impressive progress in this task, their predictions can be globally inconsistent or highly improbable. Unlike earlier methods, we treat the problem as
0
0
0
Mon May 04 2020
NLP
What-if I ask you to explain: Explaining the effects of perturbations in procedural text
QuarterET is a system that constructs explanations from paragraphs. It provides better explanations (based on the sentences in the procedural text) compared to several strong baselines on a recent process comprehension benchmark.
0
0
0
Fri Jun 21 2019
Machine Learning
Be Consistent! Improving Procedural Text Comprehension using Label Consistency
Our goal is procedural text comprehension, namely tracking how the properties of entities change with time given a procedural text. Despite recent advances, current systems still struggle with this task. Our approach is to leverage the fact that, for many procedural texts, multiple independent descriptions are readily available.
0
0
0
Thu May 17 2018
NLP
Tracking State Changes in Procedural Text: A Challenge Dataset and Models for Process Paragraph Comprehension
New dataset, ProPara, is the first to contain natural text about a changing world. The end-task, tracking the location and existence of entities through the text, is challenging because the causal effects of actions are often implicit.
0
0
0
Sun Apr 15 2018
Artificial Intelligence
What Happened? Leveraging VerbNet to Predict the Effects of Actions in Procedural Text
The goal is to answer questions about paragraphs describing processes (e.g.,photosynthesis) Texts of this genre are challenging because the effects of actions are often implicit. To supply this knowledge, we leverage VerbNet to build a rulebase (called the Semantic Lexicon)
0
0
0
Fri Aug 16 2019
NLP
Reasoning Over Paragraph Effects in Situations
A key component of successfully reading a passage of text is the ability to apply knowledge gained from the passage to a new situation. We present ROPES, a challenging benchmark for reading comprehension targeting Reasoning Over Paragraph Effects in Situations.
0
0
0